System

Abstract

A system is provided.SOLUTION: A system comprising: means for generating a question using a generative model; means for displaying the generated question on a user terminal; means for collecting a user's answer and transmitting the answer to a server; means for analyzing the answer and determining whether there is an abnormality in a cognitive function; and means for notifying a detected abnormality.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Dementia is a progressive disease, and early detection and treatment are important. However, signs of dementia are often overlooked, especially in elderly people living alone. Furthermore, daily monitoring for early detection can be a burden for the individual and their family. There is a need for a system that solves these issues and enables early detection of dementia without burdening elderly people living alone. [Means for solving the problem]

[0005] This invention provides a system that uses a generative model to generate questions on a daily basis and displays them on a user's device. Specifically, the system presents the generated questions to the user, collects the user's answers, and sends them to a server. The server then analyzes the answers using the generative AI model to determine whether there are any abnormalities in cognitive function. If an abnormality is detected, the system notifies the user and registered contacts (family members and medical institutions). Furthermore, by making the content of the questions changeable and allowing notification recipients to be selected, continuous and flexible monitoring is achieved. This enables early detection and treatment of dementia without burden, even for elderly people living alone.

[0006] A "generative model" is an algorithm that uses natural language processing and machine learning techniques to automatically generate questions suitable for dialogue with users.

[0007] The "means for generating questions" is a function for generating appropriate questions for the user using a generative model.

[0008] A "user terminal" is a device used by a user (e.g., a smartphone, tablet, or smart speaker) that displays questions and collects user responses.

[0009] The "means for collecting user responses" is a function that records responses entered by users into the terminal as text or voice data and transmits them to the server.

[0010] "Server" means a central processing unit that hosts the generative model and analyzes and stores the response data submitted by users.

[0011] The "analysis means" is a function that uses an analytical algorithm such as a generative model installed on the server to analyze the user's response data and determine whether there are any abnormalities.

[0012] "Cognitive abnormalities" refers to behavioral patterns that indicate dementia, such as inconsistent responses or memory confusion, when comparing a user's response data with past data.

[0013] The "notification means" is a function that, when the server detects an abnormality in cognitive function, sends a message to notify the user, family, and medical institutions of the occurrence of the abnormality.

[0014] "When an abnormality is detected" refers to a state in which an abnormal pattern related to cognitive function is found in the user's response data using the analysis means.

[0015] "Notification destination" refers to an individual or organization that is set in advance by the user and that will be notified when an abnormality is detected.

[0016] The "means for changing the content of a question" is a function for dynamically changing the content of a question generated by the question generating means in accordance with the user's state and past data.

[0017] "Selectable contact means" is a function that allows the notification means to select the person to whom the notification is sent based on pre-defined settings.

[0018] "Long-term data analysis" is the process of monitoring changes in a user's cognitive function over a long period of time based on data collected and accumulated by the server.

[0019] These definitions clarify important terms to facilitate understanding of the claims. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4]FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0021] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0022] First, the terms used in the following description will be explained.

[0023] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0024] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0025] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0026] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0027] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0028] [First embodiment]

[0029] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0030] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0031] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0032] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0033] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0034] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0035] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0036] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0037] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0038] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0039] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0040] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0041] The present invention is a system aimed at early detection and treatment of dementia. This system generates questions using a generative model, presents the questions to a user via a user terminal, collects the answers, and analyzes them on a server, thereby notifying the user if any abnormalities in cognitive function are detected. The detailed implementation of this system is described below.

[0042] Question generation and display

[0043] The server uses the generative model to generate appropriate questions for the user, such as "What did you eat this morning?" or "Where did you go today?" every morning. These questions are important for gathering information about the user's lifestyle and detecting changes in cognitive function.

[0044] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[0045] Collecting and sending user responses

[0046] The user answers the displayed questions naturally. For example, the user might answer, "I ate bread and coffee." This answer is recorded as text data by the user's device. The device then transmits the collected answer data to the server.

[0047] Data Analysis and Anomaly Detection

[0048] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been responding "I ate coffee and bread" for the past week, but suddenly responds "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[0049] If an anomaly is detected based on the analysis results, the server will send a notification to the configured contacts, which can include a specific response to the anomaly and a recommended course of action.

[0050] Notification and follow-up

[0051] If an abnormality is detected, the server will notify the user, their family, and medical institutions. For example, a message such as "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[0052] Continuous data collection

[0053] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[0054] This system can promote the early detection and treatment of dementia without burdening even elderly people living alone. As a concrete example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the device will send the answer to the server, which will analyze it. If any abnormalities are detected as a result, a notification will be sent. This process makes continuous monitoring possible without any strain on the user as they go about their daily lives.

[0055] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of the elderly and contribute to early treatment.

[0056] The processing flow will be explained below.

[0057] Step 1:

[0058] The server runs the generative model to generate daily questions to check the user's cognitive functions.

[0059] Step 2:

[0060] The server transmits the generated question to the user terminal.

[0061] Step 3:

[0062] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[0063] Step 4:

[0064] The user answers questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user answers "I had bread and coffee."

[0065] Step 5:

[0066] The terminal records the user's answer as text data and transmits it to the server.

[0067] Step 6:

[0068] The server uses the generative model to analyze the received user response data, utilizing natural language processing techniques.

[0069] Step 7:

[0070] The server compares the response data with a database of past responses, checking for consistency and unusual patterns.

[0071] Step 8:

[0072] If the analysis detects an abnormality in cognitive function, the server initiates a notification process.

[0073] Step 9:

[0074] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution.

[0075] Step 10:

[0076] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[0077] Step 11:

[0078] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[0079] Step 12:

[0080] The terminal repeatedly displays questions on a daily basis and continuously transmits the user's answer data to the server.

[0081] In this way, the system continuously monitors the user's cognitive function and promptly notifies them of any abnormalities, facilitating early treatment.

[0082] Example 1

[0083] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0084] Early detection of changes in cognitive function in the elderly is important, but conventional systems require cumbersome daily data collection and analysis, making continuous monitoring difficult. Furthermore, notification methods used when abnormalities are detected are uniform, making it difficult to respond to individual circumstances. Therefore, there is a need for a system that can effectively monitor changes in cognitive function and, when abnormalities are detected, can provide prompt and appropriate responses.

[0085] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0086] In this invention, the server includes means for generating questions using a generative model, means for transmitting the generated questions to an information terminal, means for displaying the questions on the information terminal, means for collecting user answers and transmitting them to a storage device, means for analyzing the answers and determining whether there is an abnormality in cognitive function, and means for notifying the user of an abnormality based on the determination result. This makes it possible to continuously monitor the user's cognitive function in their daily life, and, if an abnormality is detected, to take prompt action according to the individual situation.

[0087] A "generative model" is a machine learning algorithm that generates appropriate questions based on prompt text.

[0088] An "information terminal" is a device that allows users to display questions and input answers, and includes smartphones, tablets, smart speakers, etc.

[0089] The "storage device" is a database or storage system that stores response data sent by users and analyzes it later.

[0090] "Analysis" refers to the process of processing the collected data using natural language processing techniques and the like to evaluate the state of the user's cognitive function.

[0091] "Abnormal" is a state that indicates a case where behavior or patterns that are different from normal are observed based on the user's response data.

[0092] "Notification" refers to the act of sending a warning or advice to the user, their family, or a medical institution when an abnormality is detected.

[0093] The present invention is a system for early detection of changes in cognitive function. This system generates appropriate questions using a generative AI model, presents the questions to a user via an information terminal, collects the answers, transmits them to a storage device, and analyzes them on a server to detect abnormalities in cognitive function. Detailed implementation methods of this system are described below.

[0094] Question generation and display

[0095] The server uses a generative model to generate appropriate questions for the user. This generative model receives a prompt as input and outputs a question based on it. An example of a prompt is "Please generate appropriate questions to ask the user to check their cognitive function." The generated questions cover various aspects of the user's daily life, such as "What did you eat this morning?" and "Where did you go today?"

[0096] The generated question is sent to an information terminal via the Internet. Information terminals are devices such as smartphones, tablets, and smart speakers, and these terminals display the question to the user by voice or text. For example, in the case of a smartphone, the message "What did you have for breakfast this morning?" is displayed on the screen, and in devices that can use voice assistants, the question is read aloud.

[0097] Collecting and sending user responses

[0098] The user answers the questions displayed or read aloud on the information terminal. For example, the user answers, "I ate bread and coffee." This answer is recorded as text data by the terminal. The recorded answer data is transmitted to a storage device via the Internet. The HTTPS protocol is used as the transmission protocol to ensure data security.

[0099] Data Analysis and Anomaly Detection

[0100] The server receives and analyzes the response data sent by the user. This analysis uses natural language processing technology. Specifically, the response content is converted into a feature vector and compared with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this could be detected as an abnormality in cognitive function.

[0101] Notification and follow-up

[0102] If an abnormality is detected based on the analysis results, the server will send a notification to the specified contacts. The notification will include the specific answers that showed the abnormality and recommended actions to take. For example, a message may be sent saying, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[0103] Continuous data collection

[0104] The information terminal continues to display similar questions on a daily basis, continuously collecting data. The server stores the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[0105] As a concrete example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the answer is sent to the server, which analyzes it. If an abnormality is detected, a notification will be sent. This process makes continuous monitoring possible without any effort in everyday life.

[0106] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of elderly people and contribute to early treatment.

[0107] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0108] Step 1: Question Generation

[0109] The server inputs a prompt to the generative model: "Generate an appropriate question to ask the user to check their cognitive function." The generative model generates a question based on this prompt. For example, the output question might be, "What did you have for breakfast this morning?" This generated question is saved in a database. The data processing performed by the server involves analyzing the prompt and generating an appropriate question.

[0110] Step 2: Submit your question

[0111] The server sends the generated question to the information terminal. At this time, a protocol for transferring data over the Internet (e.g., HTTPS) is used. The input from the server to the terminal is the generated question, and the output is the state in which the question has been sent to the terminal. The server waits for an ACK (acknowledgement) to confirm that the terminal has received the question.

[0112] Step 3: Display questions

[0113] The device receives the question from the server and displays it to the user in text or voice. For example, a message such as "What did you have for breakfast this morning?" is displayed on a smartphone screen. On devices with voice output, the question is read aloud through a voice assistant. The input to the device is the question received from the server, and the output is the question displayed to the user.

[0114] Step 4: User answers

[0115] The user inputs an answer to a question displayed on the terminal. For example, the answer may be "I ate bread and coffee." This answer can be entered as text or by voice using speech recognition. The user's input is the answer to the question, and the output is the answer entered into the terminal.

[0116] Step 5: Submit your response

[0117] The terminal records the user's answer as text data and sends it to the server. The transmission is via the Internet using the HTTPS protocol. The input to the terminal is the user's answer, and the output is the answer data sent to the server. The terminal waits for an ACK (acknowledgment) to confirm that the server has received the data.

[0118] Step 6: Data analysis

[0119] The server analyzes the received response data. Natural language processing technology is used for this analysis. The response content is converted into a feature vector and compared with a database of past responses. The server's input is the response data, and the output is the analysis result. For example, if a user who answered "I ate coffee and bread" in the past week answers "I ate nothing," the analysis result will detect an abnormality in cognitive function.

[0120] Step 7: Anomaly detection and notification

[0121] If the server detects an abnormality based on the analysis results, it will send a notification to the specified contacts. The notification will include the specific answer for the abnormality and recommended countermeasures. The server's input is the analysis results, and the output is an abnormality notification. For example, a message may be sent saying, "Your answers about breakfast are inconsistent, so please consider visiting a medical institution." An ACK (acknowledgement) is waited for to confirm that the notification was sent successfully.

[0122] Step 8: Continuous data collection

[0123] The terminal continues to display questions to the user on a daily basis. For example, it displays the question "What did you eat this morning?" every morning at 8:00. The server receives new data and stores it in the database. The terminal's input is the periodically generated question, and its output is the question displayed to the user. The server's input is the new answer data, and its output is the updated database. The database is updated and optimized for the next analysis.

[0124] (Application example 1)

[0125] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0126] It is important to detect changes in cognitive function in the elderly early and take appropriate measures promptly. However, for elderly people living alone or with busy families, daily monitoring of cognitive function is a heavy burden. In addition, manual notification methods when abnormalities are detected are time-consuming and may delay appropriate responses. It is necessary to solve these issues and improve the quality of life of the elderly.

[0127] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0128] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying the user of a detected abnormality, means for detecting an abnormality by comparing the user's past answers with the user's current answers, and means for notifying the user of a detected abnormality by email. This makes it possible to continuously monitor the user's daily life, detect abnormalities in cognitive function early, and automatically notify the user.

[0129] A "generative model" is an algorithm that generates new data based on existing data. It uses AI techniques to create new questions and answers based on specific conditions and patterns.

[0130] A "user terminal" is a device operated by a user, and includes a smartphone, tablet, or PC.

[0131] A "server" is a computer system that receives data from users over a network and processes and analyzes it.

[0132] The "means for generating questions" is the process by which the server uses the generative model to generate appropriate questions for the user.

[0133] The "means for displaying on the user terminal" is a function for displaying the generated question on the screen of the user terminal.

[0134] The "means of collecting answers" is the process of acquiring the answers users give to questions as data.

[0135] "Means for sending to the server" refers to the process of sending the response data collected from the user terminals to the server via the Internet.

[0136] "Means for analyzing responses" refers to the server's ability to verify and analyze the collected response data and identify specific patterns or anomalies.

[0137] "Means for determining whether there is any abnormality in cognitive function" is a process for evaluating whether there is any change or abnormality in the user's cognitive function based on the response data.

[0138] The "means for notifying abnormalities" is a function that sends an alert to a pre-set contact when an abnormality is detected.

[0139] "Means for comparing past answers with current answers" refers to the process of comparing a user's answer history with their most recent answers to check for consistency or anomalies.

[0140] "Means for notifying by email" is a function that automatically sends detected abnormalities to relevant parties by email.

[0141] The present invention is a system for early detection of changes in cognitive function and providing appropriate notifications when abnormalities are detected. This system generates questions using a generative model, presents the questions to the user via a user terminal, collects the answers, and analyzes them on a server, thereby providing notifications when abnormalities are detected in cognitive function.

[0142] Question generation and display

[0143] The server uses the generative model to generate appropriate questions. For example, questions such as "What did you eat this morning?" or "Where did you go today?" are generated. These questions are important for collecting information about the user's daily life. The generated questions are sent via the Internet to a user device such as a smartphone. The user device displays the questions on a screen and, in some cases, presents them to the user by voice.

[0144] Collecting and sending user responses

[0145] The user responds naturally to the displayed questions. For example, if the user responds "I ate bread and coffee," the device records the response as text data. The device then transmits the collected response data to the server.

[0146] Data Analysis and Anomaly Detection

[0147] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[0148] Abnormal notification

[0149] If an abnormality is detected based on the analysis results, the server will send a notification to pre-defined contacts. The notification will include the specific abnormality detected and recommended action to take. Notifications are automatically sent via email, allowing users, their families, and medical institutions to take action quickly.

[0150] Continuous data collection

[0151] The user's device continues to display similar questions on a daily basis, collecting data continuously. The server receives new data, stores it in a database, and optimizes it for the next analysis. This makes it possible to analyze data over a long period of time, enabling highly accurate monitoring of changes in the user's cognitive function.

[0152] Specific examples

[0153] For example, consider a case where a user answers the question "What did you eat this morning?" every morning. Suppose a user who has answered "I had bread and coffee" for a week in a row suddenly answers "I didn't eat anything." If this abnormal answer is detected, the server automatically sends an email to a pre-set contact with a message stating, "An abnormality has been detected. Your recent answers have been inconsistent." This system can constantly monitor the health of elderly people, detect abnormalities early, and promote appropriate measures.

[0154] Prompt Sentence Examples

[0155] What did you eat this morning?

[0156] What did you have for dinner last night?

[0157] Where are you planning to go today?

[0158] How are you feeling right now?

[0159] The above is an embodiment of the present invention. This system is extremely effective for efficiently monitoring the cognitive function of elderly people and promoting early treatment.

[0160] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0161] Step 1:

[0162] The server generates questions using a generative AI model.

[0163] Input: Past question data and user profile information.

[0164] Data processing: The generative AI model selects the best questions based on the conditions and generates new questions.

[0165] Output: The generated question.

[0166] Step 2:

[0167] The server transmits the generated question to the user terminal.

[0168] Input: The generated question.

[0169] Data processing: The question is converted into a data packet for transmission and sent over the Internet.

[0170] Output: The question is displayed on the user's terminal.

[0171] Step 3:

[0172] The user answers the questions displayed.

[0173] Input: The question displayed to the user.

[0174] Data processing: The user enters the answer, and the answer is recorded as text data.

[0175] Output: User answers as text data.

[0176] Step 4:

[0177] The user terminal transmits the collected response data to the server.

[0178] Input: User response text data.

[0179] Data processing: The response data is converted into a data packet for transmission and sent to the server.

[0180] Output: The response data sent to the server.

[0181] Step 5:

[0182] The server receives the response data sent by the user and begins analyzing it.

[0183] Input: The response data sent to the server.

[0184] Data processing: Using natural language processing techniques, responses are converted into feature vectors and compared with a database of past responses.

[0185] Output: Data analysis results.

[0186] Step 6:

[0187] The server compares past responses with current responses to determine whether there are any abnormalities in cognitive function.

[0188] Input: Past response database and current response data.

[0189] Data processing: Comparative analysis of the consistency of answers and abnormal patterns.

[0190] Output: Abnormality judgment result.

[0191] Step 7:

[0192] The server generates and emails notifications if any anomalies are detected.

[0193] Input: Abnormality detection result and pre-set notification destination information.

[0194] Data processing: A notification containing the details of the abnormality and countermeasures is created, converted into email format, and sent.

[0195] Output: Email notification.

[0196] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0197] This invention is a system aimed at early detection and treatment of dementia, and combines question generation using a generative model, collection and analysis of user responses, and emotion recognition using an emotion engine. The detailed implementation method of this system is described below.

[0198] Question generation and display

[0199] The server runs the generative model to generate everyday questions to check the user's cognitive function. For example, every morning it generates questions such as "What did you eat this morning?" and "How was your day?" These questions are important for collecting information about the user's lifestyle and detecting changes in cognitive function.

[0200] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[0201] Collecting and sending user responses

[0202] The user responds naturally to the displayed question. For example, suppose the user responds, "I ate bread and coffee." This response is recorded as text data by the user's device, and the emotion engine simultaneously analyzes the user's emotions. The device then transmits the collected response data and emotion data to the server.

[0203] Data Analysis and Anomaly Detection

[0204] The server receives the response data and emotion data sent by the user and analyzes the content using a generative model. Natural language processing technology is used for the analysis. The response content is converted into a feature vector and compared with a database of past responses. Meanwhile, the emotion data analyzed by the emotion engine is also added to the anomaly detection process.

[0205] For example, if a user has answered "I ate coffee and bread" over the past week and always displayed a calm emotion, but suddenly answers sadly and says "I haven't eaten anything," this would be judged to be a cognitive abnormality.

[0206] Notification and follow-up

[0207] If the analysis detects any abnormalities in cognitive function, the server initiates the notification process. It generates a notification containing the abnormality and its details and sends it to registered contacts. Recipients of the notification include the user, their family, and medical institutions.

[0208] For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a doctor. Also, based on your emotional data, it's likely that you're feeling anxious."

[0209] Continuous data collection

[0210] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. It also accumulates emotional data to improve the accuracy of anomaly detection.

[0211] Specific examples

[0212] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. After the server analyzes the data and compares it with past data, if it determines that there are no abnormalities, it will be stored in the database as is. On the other hand, if an abnormality is detected, an appropriate notification will be sent to the user and necessary contacts.

[0213] This concludes the details of the embodiment of the present invention. This system enables advanced monitoring that takes into account not only cognitive function but also the user's emotional state, improving the quality of life for the elderly and contributing to early treatment.

[0214] The processing flow will be explained below.

[0215] Step 1:

[0216] The server runs a generative model to generate everyday questions, such as "What did you have for breakfast this morning?" or "How was your day?"

[0217] Step 2:

[0218] The server transmits the generated question to the user terminal.

[0219] Step 3:

[0220] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[0221] Step 4:

[0222] The user responds naturally to questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user responds "I had bread and coffee."

[0223] Step 5:

[0224] The device records the user's response as text data and simultaneously uses an emotion engine to analyze the user's emotion when responding, for example, by recognizing emotion from voice tone.

[0225] Step 6:

[0226] The terminal transmits the collected response data and emotion data to the server.

[0227] Step 7:

[0228] The server uses the generative model to analyze the received user response data, using natural language processing techniques.

[0229] Step 8:

[0230] The server compares the response data with a database of past responses to check for consistency and unusual patterns, and also combines and analyzes the emotional data to identify any anomalies.

[0231] Step 9:

[0232] If the server detects an anomaly in the response data or emotion data as a result of the analysis, it initiates a notification process.

[0233] Step 10:

[0234] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution. For example, a message saying, "Your recent breakfast answers have been inconsistent, so please consider seeking medical advice. You may also be feeling anxious."

[0235] Step 11:

[0236] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[0237] Step 12:

[0238] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[0239] Step 13:

[0240] The terminal repeatedly asks similar questions on a daily basis and continuously collects data on the user's responses.

[0241] This system improves the accuracy of detecting abnormalities in cognitive function by taking the user's emotional state into consideration. As a specific example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning, and User A will respond with "I had tamagoyaki and rice today." If the emotion engine detects the emotion "calm," the response data and emotion data will be sent to the server. If the server's analysis reveals inconsistencies or changes in emotion, it will send an appropriate notification. In this way, continuous monitoring can be performed effortlessly in everyday life.

[0242] Example 2

[0243] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0244] As the aging society advances, early detection and treatment of dementia are becoming increasingly important. However, current methods for detecting cognitive function do not adequately reflect changes in the user's daily life or emotional state, making effective monitoring impossible. This results in low accuracy in detecting abnormalities, making it difficult to improve the user's quality of life and realize early treatment.

[0245] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0246] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for analyzing emotions contained in the answers using an emotion engine, means for detecting an abnormality based on the analysis result and emotion data, and means for notifying of detected abnormalities. This enables advanced monitoring that takes into account not only cognitive function but also emotional states.

[0247] A "generative model" is an artificial intelligence algorithm that uses natural language processing techniques to generate text for a specific task.

[0248] A "terminal" is a hardware device that can be directly operated by a user, and includes smartphones, tablets, smart speakers, etc.

[0249] A "server" is a central computer system that communicates with terminals via a network and processes and manages data.

[0250] The "means for generating questions" is a function that uses a generative model to automatically create questions to verify the user's cognitive functions.

[0251] The "means for displaying a question on a terminal" is a function for transmitting the generated question to a terminal via the Internet and presenting it to the user on that terminal in voice or text format.

[0252] The "means for collecting answers and sending them to a server" is a function for collecting answers entered by users to questions and sending them to a server via a network.

[0253] "Means for analyzing responses and determining whether there are any abnormalities in cognitive function" refers to a function that analyzes collected response data using natural language processing technology and compares it with past response history to detect abnormalities in cognitive function.

[0254] An "emotion engine" is a software component that analyzes a user's emotional state from their text data and classifies it into specific emotional categories.

[0255] "Means for analyzing emotions" is a function that uses an emotion engine to analyze the emotions contained in the user's response and uses that information in the discrimination process.

[0256] "Means for detecting abnormalities" refers to a function that detects abnormalities in cognitive function based on analysis results and emotional data.

[0257] The "notification means" is a function that notifies designated contacts such as the user, their family, or a medical institution based on detected abnormality information.

[0258] This invention is a system aimed at early detection and treatment of dementia, and is realized by combining question generation using a generative AI model, collection and analysis of user responses, and emotion recognition using an emotion engine. An embodiment of this system is described in detail below.

[0259] Question generation and display

[0260] The server runs a generative AI model to generate everyday questions to check the user's cognitive function. The generative model can be, for example, GPT-3, a large-scale natural language processing technology. For example, it generates questions such as "What did you eat this morning?" or "How was your day?" These questions are important for collecting information about the user's lifestyle habits and detecting changes in cognitive function. The generated questions are sent via the internet to the user's device. The user's device can be a smartphone, tablet, or smart speaker, and the questions are displayed in voice or text format.

[0261] Collecting user responses and analyzing sentiment

[0262] The user responds naturally to the generated questions. The user's device records the response as text data. For example, if the user responds "I ate bread and coffee," this response text is saved on the device. The device then uses an emotion engine to analyze the emotions contained in the response data. This emotion analysis can be performed using, for example, Microsoft Azure's emotion analysis API. The analysis results are output as emotion categories such as "calm," "sad," and "happy."

[0263] Data transmission and analysis

[0264] The device sends the collected response data and emotion data to the server, where it is added to a queue for analysis. The server then uses a generative model to analyze the received data. Natural language processing techniques are used for the analysis, converting the response content into a feature vector and comparing it with a database of past responses. For example, the BERT model is used to tokenize the response and generate a feature vector. Emotion data is also analyzed at the same time to determine whether there are any anomalies.

[0265] Anomaly Detection and Notification

[0266] The server detects anomalies based on the analysis results and emotional data. It compares the results with past data to determine whether there are any abnormalities in the user's cognitive function. For example, if a user has always responded "I ate coffee and bread" over the past week, always showing a calm emotion, but suddenly responds sadly, "I haven't eaten anything," this is deemed to be an abnormality in cognitive function. If an abnormality is detected, the server initiates a notification process and sends a notification to the user, their family, and registered contacts, such as medical institutions. For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution. Also, based on your emotional data, it is likely that you are feeling anxious."

[0267] Continuous data collection

[0268] The user's device continues to display similar questions on a daily basis, continuously collecting answer data and emotional data. The server receives new data, stores it in a database, and optimizes it for the next analysis. This improves the accuracy of anomaly detection and enables long-term monitoring of the user's cognitive function and emotional state.

[0269] Specific examples

[0270] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. The server analyzes the data and, if it is determined to be normal by comparing it with past data, it is stored in the database as is. If an abnormality is detected, an appropriate notification is sent to the user and necessary contacts.

[0271] This system enables advanced monitoring that takes into account not only cognitive function but also emotional state, improving the quality of life for the elderly and contributing to early treatment.

[0272] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0273] Understood. Below, we will explain the program processing flow of the dementia early detection system by dividing it into specific processing steps.

[0274] Program processing flow

[0275] Step 1:

[0276] The server uses a generative AI model to generate questions. Specifically, it uses natural language processing techniques such as GPT-3 to create everyday questions such as "What did you have for breakfast this morning?" or "How was your day?" The input is a prompt to the generative AI model, and the output is the generated question.

[0277] Step 2:

[0278] The server sends the generated question to the user's device via the Internet. The input here is the generated question, and the output is the transmission of the question to the user's device. Specifically, the server uses an HTTP request to send the question data to a smartphone or tablet.

[0279] Step 3:

[0280] The terminal displays the received question to the user in voice or text format. The input is the question sent from the server, and the output is the user's visual or auditory recognition of the question. Specifically, the text is displayed on the terminal's display, or the question is read aloud using speech synthesis.

[0281] Step 4:

[0282] The user responds naturally to the displayed question. The input is the displayed question, and the output is the user's response in text or voice. Specifically, the user can type "I ate bread and coffee" on the smartphone keyboard, or respond vocally using the voice recognition function.

[0283] Step 5:

[0284] The device records the user's responses as text data and uses an emotion engine to analyze the emotions contained in the responses. The input is the user's response data, and the output is the emotion analysis results. Specifically, the device uses Microsoft Azure's emotion analysis API to classify responses into emotion categories such as "calm," "sad," and "happy."

[0285] Step 6:

[0286] The terminal sends the collected response data and emotion data to the server. The input is response data and emotion data, and the output is data sent to the server. Specifically, an HTTP request is generated and the response data and emotion data are sent to the server in JSON format.

[0287] Step 7:

[0288] The server adds the received data to a queue for analysis and analyzes the answer using a generative model. The input is the submitted answer data and emotion data, and the output is the analysis result. Specifically, the server uses the BERT model to tokenize the answer, convert it into a feature vector, and then analyzes it.

[0289] Step 8:

[0290] The server detects anomalies based on the analysis results and emotion data. The input is the analysis results and emotion data, and the output is whether anomalies are detected. Specifically, if anomalies are found in the response patterns or emotions by comparing them with past data, a flag is set to indicate an anomaly.

[0291] Step 9:

[0292] If an anomaly is detected, the server initiates a notification process. The input is an anomaly detection flag and related information, and the output is a notification message. Specifically, a notification is sent to the user, their family, and medical institutions stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a medical institution."

[0293] Step 10:

[0294] The terminal continues to display similar questions on a daily basis and continuously collects data. The input is new questions from the server, and the output is updated answer data. Specifically, it displays newly generated questions to the user and continues to collect answers.

[0295] The above is the specific program processing flow of the dementia early detection system. Through this detailed process, it is possible to monitor the user's daily life and emotional state, detect abnormalities early, and provide appropriate notifications.

[0296] (Application example 2)

[0297] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0298] In today's world, early detection of dementia in the elderly and appropriate treatment are becoming increasingly important as the population ages. However, many elderly people do not have the opportunity to undergo regular medical checkups, making it difficult to detect cognitive decline early. Furthermore, cognitive decline is often accompanied by emotional changes, creating a need for effective monitoring methods that take this into account. Therefore, the present invention aims to provide a system that monitors changes in cognitive function in daily life and detects abnormalities early.

[0299] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0300] In this invention, the server includes means for generating questions using a generative AI model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying of detected abnormalities, means for analyzing user emotions, and means for determining abnormalities with greater accuracy based on the emotion analysis results. This enables early detection of abnormalities in cognitive function while also enabling advanced monitoring that takes emotional changes into account.

[0301] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to automatically generate questions to present to users.

[0302] A "user terminal" is a device such as a smartphone, tablet, or smart glasses that a user can directly operate to answer questions.

[0303] A "server" is a computer system that receives data sent by users, analyzes it, and stores or uses the results.

[0304] The "answer analysis means" refers to a method or device for analyzing answer data obtained from a user and determining whether there is any abnormality in cognitive function.

[0305] The "abnormality notification means" is a method or device for sending a warning to a preset notification destination when an abnormality is detected based on the analysis results.

[0306] "Emotion analysis means" refers to a method or device for analyzing a user's emotional state based on the user's responses and related data.

[0307] The "anomaly detection means" is a method or device for detecting anomalies with higher accuracy, taking into account the results of emotion analysis.

[0308] This invention relates to a system for monitoring cognitive function and recognizing emotions of elderly customers by store staff. The system generates questions using a generative AI model, displays the questions on the user's device, and collects and analyzes the user's answers. The detailed implementation of the system is described below.

[0309] Question generation and display

[0310] The server runs a generative AI model to generate cognitive function check questions for the customer. This generation process uses the "transformers" library to generate questions by inputting prompts into the GPT-2 model. For example, the prompts might look like this:

[0311] Example prompt:

[0312] "Generate questions for your customers to help detect dementia early."

[0313] The generated questions are displayed on the smartphones or smart glasses used by store staff, who then ask these questions to customers to collect everyday information.

[0314] Collecting user responses and analyzing sentiment

[0315] The user, i.e., the customer, answers questions posed by the staff. These answers are recorded on the user's device and then subjected to emotional analysis. The emotional analysis process utilizes an emotion engine API to analyze the user's emotional state from the text of their answers.

[0316] For example, if a customer answers the question "What did you eat today?" with "I had tamagoyaki and rice today," the emotion engine can detect the emotion "calm."

[0317] Data transmission and analysis

[0318] The collected response data and emotion data are sent from the user's device to a server. The server analyzes this data and determines whether there are any abnormalities in the user's cognitive function. By taking into account not only the response data but also the emotion data, the accuracy of abnormality detection is improved.

[0319] Notification and follow-up

[0320] If the server detects an abnormality as a result of the analysis, it will send a warning to staff or caregivers via an abnormality notification method. This notification will include the nature of the abnormality and its details, and provide specific advice, such as "Your answers about your recent breakfast have been inconsistent. Please consider seeking medical advice."

[0321] Continuous data collection

[0322] The user device continues to generate similar questions on a daily basis, collecting data continuously. The server accumulates the new data it receives and optimizes it for the next analysis. Accumulating emotion data can further improve the accuracy of anomaly detection.

[0323] In this way, the present invention enables store staff to effectively monitor cognitive function and recognize emotions through interactions with elderly customers, thereby enabling early detection of cognitive decline and appropriate follow-up.

[0324] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0325] Step 1: Question Generation

[0326] The server runs the generative AI model to generate questions to present to the user. In this case, it uses the prompt "Please generate questions for the customer to help with the early detection of dementia." The input is this prompt, and the output is the generated question. The generative AI model uses a transformer model such as GPT-2 to generate natural-looking questions based on the prompt.

[0327] Step 2: View the question

[0328] The terminal receives the generated question sent from the server and displays it to the user in text or audio. The input is the generated question sent from the server, and the output is the question presented to the user. For example, the question may be displayed on a smartphone or smart glasses.

[0329] Step 3: Collect responses

[0330] The user, i.e., the customer, answers questions displayed on the terminal. The input is the displayed question and the user's answer, and the output is the answer recorded in text format. The terminal records the user's answer as text data and temporarily saves it.

[0331] Step 4: Sentiment Analysis

[0332] The device sends the collected user responses to the sentiment analysis engine API, which analyzes the emotional state of the responses. The input is the user's response as text data, and the output is data indicating the emotional state. The specific operation is to send the response text to the API and receive a label such as "calm," "sad," or "happy" as the emotional state.

[0333] Step 5: Send data

[0334] The device sends the text data and the emotion analysis results to the server. The input is the user's response and the emotion analysis results, and the output is the data sent to the server. The server records the received data in a database.

[0335] Step 6: Data analysis

[0336] The server analyzes the received response data and emotional data, and compares it with past data to determine whether there are any abnormalities in cognitive function. The input is the response data and emotional data, and the output is the analysis result (normal or abnormal). Specifically, it uses natural language processing technology to convert the response content into a feature vector, which is then compared with past data to detect abnormalities.

[0337] Step 7: Notification of abnormalities

[0338] If the server detects an anomaly, it starts a notification process and notifies registered contacts that an anomaly has been detected. The input is the analysis result (anomaly) and contact information, and the output is the notification message sent. For example, a message saying "Your answers about your recent breakfast have been inconsistent, so please consider visiting a medical institution." is sent.

[0339] Step 8: Continuous data collection

[0340] The device periodically generates similar questions, collects the user's answers, and sends them to the server. The input is the new questions and the user's new answers, and the output is a continuously updated database. The server accumulates the new data and uses it for the next analysis.

[0341] The above are the processing steps for specifically implementing cognitive function monitoring and emotion recognition for elderly customers in a physical store.

[0342] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0343] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0344] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0345] [Second embodiment]

[0346] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0347] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0348] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0349] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0350] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0351] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0352] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0353] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0354] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0355] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0356] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0357] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0358] The present invention is a system aimed at early detection and treatment of dementia. This system generates questions using a generative model, presents the questions to a user via a user terminal, collects the answers, and analyzes them on a server, thereby notifying the user if any abnormalities in cognitive function are detected. The detailed implementation of this system is described below.

[0359] Question generation and display

[0360] The server uses the generative model to generate appropriate questions for the user, such as "What did you eat this morning?" or "Where did you go today?" every morning. These questions are important for gathering information about the user's lifestyle and detecting changes in cognitive function.

[0361] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[0362] Collecting and sending user responses

[0363] The user answers the displayed questions naturally. For example, the user might answer, "I ate bread and coffee." This answer is recorded as text data by the user's device. The device then transmits the collected answer data to the server.

[0364] Data Analysis and Anomaly Detection

[0365] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been responding "I ate coffee and bread" for the past week, but suddenly responds "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[0366] If an anomaly is detected based on the analysis results, the server will send a notification to the configured contacts, which can include a specific response to the anomaly and a recommended course of action.

[0367] Notification and follow-up

[0368] If an abnormality is detected, the server will notify the user, their family, and medical institutions. For example, a message such as "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[0369] Continuous data collection

[0370] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[0371] This system can promote the early detection and treatment of dementia without burdening even elderly people living alone. As a concrete example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the device will send the answer to the server, which will analyze it. If any abnormalities are detected as a result, a notification will be sent. This process makes continuous monitoring possible without any strain on the user as they go about their daily lives.

[0372] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of the elderly and contribute to early treatment.

[0373] The processing flow will be explained below.

[0374] Step 1:

[0375] The server runs the generative model to generate daily questions to check the user's cognitive functions.

[0376] Step 2:

[0377] The server transmits the generated question to the user terminal.

[0378] Step 3:

[0379] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[0380] Step 4:

[0381] The user answers questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user answers "I had bread and coffee."

[0382] Step 5:

[0383] The terminal records the user's answer as text data and transmits it to the server.

[0384] Step 6:

[0385] The server uses the generative model to analyze the received user response data, utilizing natural language processing techniques.

[0386] Step 7:

[0387] The server compares the response data with a database of past responses, checking for consistency and unusual patterns.

[0388] Step 8:

[0389] If the analysis detects an abnormality in cognitive function, the server initiates a notification process.

[0390] Step 9:

[0391] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution.

[0392] Step 10:

[0393] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[0394] Step 11:

[0395] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[0396] Step 12:

[0397] The terminal repeatedly displays questions on a daily basis and continuously transmits the user's answer data to the server.

[0398] In this way, the system continuously monitors the user's cognitive function and promptly notifies them of any abnormalities, facilitating early treatment.

[0399] Example 1

[0400] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0401] Early detection of changes in cognitive function in the elderly is important, but conventional systems require cumbersome daily data collection and analysis, making continuous monitoring difficult. Furthermore, notification methods used when abnormalities are detected are uniform, making it difficult to respond to individual circumstances. Therefore, there is a need for a system that can effectively monitor changes in cognitive function and, when abnormalities are detected, can provide prompt and appropriate responses.

[0402] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0403] In this invention, the server includes means for generating questions using a generative model, means for transmitting the generated questions to an information terminal, means for displaying the questions on the information terminal, means for collecting user answers and transmitting them to a storage device, means for analyzing the answers and determining whether there is an abnormality in cognitive function, and means for notifying the user of an abnormality based on the determination result. This makes it possible to continuously monitor the user's cognitive function in their daily life, and, if an abnormality is detected, to take prompt action according to the individual situation.

[0404] A "generative model" is a machine learning algorithm that generates appropriate questions based on prompt text.

[0405] An "information terminal" is a device that allows users to display questions and input answers, and includes smartphones, tablets, smart speakers, etc.

[0406] The "storage device" is a database or storage system that stores response data sent by users and analyzes it later.

[0407] "Analysis" refers to the process of processing the collected data using natural language processing techniques and the like to evaluate the state of the user's cognitive function.

[0408] "Abnormal" is a state that indicates a case where behavior or patterns that are different from normal are observed based on the user's response data.

[0409] "Notification" refers to the act of sending a warning or advice to the user, their family, or a medical institution when an abnormality is detected.

[0410] The present invention is a system for early detection of changes in cognitive function. This system generates appropriate questions using a generative AI model, presents the questions to a user via an information terminal, collects the answers, transmits them to a storage device, and analyzes them on a server to detect abnormalities in cognitive function. Detailed implementation methods of this system are described below.

[0411] Question generation and display

[0412] The server uses a generative model to generate appropriate questions for the user. This generative model receives a prompt as input and outputs a question based on it. An example of a prompt is "Please generate appropriate questions to ask the user to check their cognitive function." The generated questions cover various aspects of the user's daily life, such as "What did you eat this morning?" and "Where did you go today?"

[0413] The generated question is sent to an information terminal via the Internet. Information terminals are devices such as smartphones, tablets, and smart speakers, and these terminals display the question to the user by voice or text. For example, in the case of a smartphone, the message "What did you have for breakfast this morning?" is displayed on the screen, and in devices that can use voice assistants, the question is read aloud.

[0414] Collecting and sending user responses

[0415] The user answers the questions displayed or read aloud on the information terminal. For example, the user answers, "I ate bread and coffee." This answer is recorded as text data by the terminal. The recorded answer data is transmitted to a storage device via the Internet. The HTTPS protocol is used as the transmission protocol to ensure data security.

[0416] Data Analysis and Anomaly Detection

[0417] The server receives and analyzes the response data sent by the user. This analysis uses natural language processing technology. Specifically, the response content is converted into a feature vector and compared with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this could be detected as an abnormality in cognitive function.

[0418] Notification and follow-up

[0419] If an abnormality is detected based on the analysis results, the server will send a notification to the specified contacts. The notification will include the specific answers that showed the abnormality and recommended actions to take. For example, a message may be sent saying, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[0420] Continuous data collection

[0421] The information terminal continues to display similar questions on a daily basis, continuously collecting data. The server stores the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[0422] As a concrete example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the answer is sent to the server, which analyzes it. If an abnormality is detected, a notification will be sent. This process makes continuous monitoring possible without any effort in everyday life.

[0423] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of elderly people and contribute to early treatment.

[0424] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0425] Step 1: Question Generation

[0426] The server inputs a prompt to the generative model: "Generate an appropriate question to ask the user to check their cognitive function." The generative model generates a question based on this prompt. For example, the output question might be, "What did you have for breakfast this morning?" This generated question is saved in a database. The data processing performed by the server involves analyzing the prompt and generating an appropriate question.

[0427] Step 2: Submit your question

[0428] The server sends the generated question to the information terminal. At this time, a protocol for transferring data over the Internet (e.g., HTTPS) is used. The input from the server to the terminal is the generated question, and the output is the state in which the question has been sent to the terminal. The server waits for an ACK (acknowledgement) to confirm that the terminal has received the question.

[0429] Step 3: Display questions

[0430] The device receives the question from the server and displays it to the user in text or voice. For example, a message such as "What did you have for breakfast this morning?" is displayed on a smartphone screen. On devices with voice output, the question is read aloud through a voice assistant. The input to the device is the question received from the server, and the output is the question displayed to the user.

[0431] Step 4: User answers

[0432] The user inputs an answer to a question displayed on the terminal. For example, the answer may be "I ate bread and coffee." This answer can be entered as text or by voice using speech recognition. The user's input is the answer to the question, and the output is the answer entered into the terminal.

[0433] Step 5: Submit your response

[0434] The terminal records the user's answer as text data and sends it to the server. The transmission is via the Internet using the HTTPS protocol. The input to the terminal is the user's answer, and the output is the answer data sent to the server. The terminal waits for an ACK (acknowledgment) to confirm that the server has received the data.

[0435] Step 6: Data analysis

[0436] The server analyzes the received response data. Natural language processing technology is used for this analysis. The response content is converted into a feature vector and compared with a database of past responses. The server's input is the response data, and the output is the analysis result. For example, if a user who answered "I ate coffee and bread" in the past week answers "I ate nothing," the analysis result will detect an abnormality in cognitive function.

[0437] Step 7: Anomaly detection and notification

[0438] If the server detects an abnormality based on the analysis results, it will send a notification to the specified contacts. The notification will include the specific answer for the abnormality and recommended countermeasures. The server's input is the analysis results, and the output is an abnormality notification. For example, a message may be sent saying, "Your answers about breakfast are inconsistent, so please consider visiting a medical institution." An ACK (acknowledgement) is waited for to confirm that the notification was sent successfully.

[0439] Step 8: Continuous data collection

[0440] The terminal continues to display questions to the user on a daily basis. For example, it displays the question "What did you eat this morning?" every morning at 8:00. The server receives new data and stores it in the database. The terminal's input is the periodically generated question, and its output is the question displayed to the user. The server's input is the new answer data, and its output is the updated database. The database is updated and optimized for the next analysis.

[0441] (Application example 1)

[0442] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0443] It is important to detect changes in cognitive function in the elderly early and take appropriate measures promptly. However, for elderly people living alone or with busy families, daily monitoring of cognitive function is a heavy burden. In addition, manual notification methods when abnormalities are detected are time-consuming and may delay appropriate responses. It is necessary to solve these issues and improve the quality of life of the elderly.

[0444] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0445] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying the user of a detected abnormality, means for detecting an abnormality by comparing the user's past answers with the user's current answers, and means for notifying the user of a detected abnormality by email. This makes it possible to continuously monitor the user's daily life, detect abnormalities in cognitive function early, and automatically notify the user.

[0446] A "generative model" is an algorithm that generates new data based on existing data. It uses AI techniques to create new questions and answers based on specific conditions and patterns.

[0447] A "user terminal" is a device operated by a user, and includes a smartphone, tablet, or PC.

[0448] A "server" is a computer system that receives data from users over a network and processes and analyzes it.

[0449] The "means for generating questions" is the process by which the server uses the generative model to generate appropriate questions for the user.

[0450] The "means for displaying on the user terminal" is a function for displaying the generated question on the screen of the user terminal.

[0451] The "means of collecting answers" is the process of acquiring the answers users give to questions as data.

[0452] "Means for sending to the server" refers to the process of sending the response data collected from the user terminals to the server via the Internet.

[0453] "Means for analyzing responses" refers to the server's ability to verify and analyze the collected response data and identify specific patterns or anomalies.

[0454] "Means for determining whether there is any abnormality in cognitive function" is a process for evaluating whether there is any change or abnormality in the user's cognitive function based on the response data.

[0455] The "means for notifying abnormalities" is a function that sends an alert to a pre-set contact when an abnormality is detected.

[0456] "Means for comparing past answers with current answers" refers to the process of comparing a user's answer history with their most recent answers to check for consistency or anomalies.

[0457] "Means for notifying by email" is a function that automatically sends detected abnormalities to relevant parties by email.

[0458] The present invention is a system for early detection of changes in cognitive function and providing appropriate notifications when abnormalities are detected. This system generates questions using a generative model, presents the questions to the user via a user terminal, collects the answers, and analyzes them on a server, thereby providing notifications when abnormalities are detected in cognitive function.

[0459] Question generation and display

[0460] The server uses the generative model to generate appropriate questions. For example, questions such as "What did you eat this morning?" or "Where did you go today?" are generated. These questions are important for collecting information about the user's daily life. The generated questions are sent via the Internet to a user device such as a smartphone. The user device displays the questions on a screen and, in some cases, presents them to the user by voice.

[0461] Collecting and sending user responses

[0462] The user responds naturally to the displayed questions. For example, if the user responds "I ate bread and coffee," the device records the response as text data. The device then transmits the collected response data to the server.

[0463] Data Analysis and Anomaly Detection

[0464] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[0465] Abnormal notification

[0466] If an abnormality is detected based on the analysis results, the server will send a notification to pre-defined contacts. The notification will include the specific abnormality detected and recommended action to take. Notifications are automatically sent via email, allowing users, their families, and medical institutions to take action quickly.

[0467] Continuous data collection

[0468] The user's device continues to display similar questions on a daily basis, collecting data continuously. The server receives new data, stores it in a database, and optimizes it for the next analysis. This makes it possible to analyze data over a long period of time, enabling highly accurate monitoring of changes in the user's cognitive function.

[0469] Specific examples

[0470] For example, consider a case where a user answers the question "What did you eat this morning?" every morning. Suppose a user who has answered "I had bread and coffee" for a week in a row suddenly answers "I didn't eat anything." If this abnormal answer is detected, the server automatically sends an email to a pre-set contact with a message stating, "An abnormality has been detected. Your recent answers have been inconsistent." This system can constantly monitor the health of elderly people, detect abnormalities early, and promote appropriate measures.

[0471] Prompt Sentence Examples

[0472] What did you eat this morning?

[0473] What did you have for dinner last night?

[0474] Where are you planning to go today?

[0475] How are you feeling right now?

[0476] The above is an embodiment of the present invention. This system is extremely effective for efficiently monitoring the cognitive function of elderly people and promoting early treatment.

[0477] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0478] Step 1:

[0479] The server generates questions using a generative AI model.

[0480] Input: Past question data and user profile information.

[0481] Data processing: The generative AI model selects the best questions based on the conditions and generates new questions.

[0482] Output: The generated question.

[0483] Step 2:

[0484] The server transmits the generated question to the user terminal.

[0485] Input: The generated question.

[0486] Data processing: The question is converted into a data packet for transmission and sent over the Internet.

[0487] Output: The question is displayed on the user's terminal.

[0488] Step 3:

[0489] The user answers the questions displayed.

[0490] Input: The question displayed to the user.

[0491] Data processing: The user enters the answer, and the answer is recorded as text data.

[0492] Output: User answers as text data.

[0493] Step 4:

[0494] The user terminal transmits the collected response data to the server.

[0495] Input: User response text data.

[0496] Data processing: The response data is converted into a data packet for transmission and sent to the server.

[0497] Output: The response data sent to the server.

[0498] Step 5:

[0499] The server receives the response data sent by the user and begins analyzing it.

[0500] Input: The response data sent to the server.

[0501] Data processing: Using natural language processing techniques, responses are converted into feature vectors and compared with a database of past responses.

[0502] Output: Data analysis results.

[0503] Step 6:

[0504] The server compares past responses with current responses to determine whether there are any abnormalities in cognitive function.

[0505] Input: Past response database and current response data.

[0506] Data processing: Comparative analysis of the consistency of answers and abnormal patterns.

[0507] Output: Abnormality judgment result.

[0508] Step 7:

[0509] The server generates and emails notifications if any anomalies are detected.

[0510] Input: Abnormality detection result and pre-set notification destination information.

[0511] Data processing: A notification containing the details of the abnormality and countermeasures is created, converted into email format, and sent.

[0512] Output: Email notification.

[0513] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0514] This invention is a system aimed at early detection and treatment of dementia, and combines question generation using a generative model, collection and analysis of user responses, and emotion recognition using an emotion engine. The detailed implementation method of this system is described below.

[0515] Question generation and display

[0516] The server runs the generative model to generate everyday questions to check the user's cognitive function. For example, every morning it generates questions such as "What did you eat this morning?" and "How was your day?" These questions are important for collecting information about the user's lifestyle and detecting changes in cognitive function.

[0517] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[0518] Collecting and sending user responses

[0519] The user responds naturally to the displayed question. For example, suppose the user responds, "I ate bread and coffee." This response is recorded as text data by the user's device, and the emotion engine simultaneously analyzes the user's emotions. The device then transmits the collected response data and emotion data to the server.

[0520] Data Analysis and Anomaly Detection

[0521] The server receives the response data and emotion data sent by the user and analyzes the content using a generative model. Natural language processing technology is used for the analysis. The response content is converted into a feature vector and compared with a database of past responses. Meanwhile, the emotion data analyzed by the emotion engine is also added to the anomaly detection process.

[0522] For example, if a user has answered "I ate coffee and bread" over the past week and always displayed a calm emotion, but suddenly answers sadly and says "I haven't eaten anything," this would be judged to be a cognitive abnormality.

[0523] Notification and follow-up

[0524] If the analysis detects any abnormalities in cognitive function, the server initiates the notification process. It generates a notification containing the abnormality and its details and sends it to registered contacts. Recipients of the notification include the user, their family, and medical institutions.

[0525] For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a doctor. Also, based on your emotional data, it's likely that you're feeling anxious."

[0526] Continuous data collection

[0527] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. It also accumulates emotional data to improve the accuracy of anomaly detection.

[0528] Specific examples

[0529] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. After the server analyzes the data and compares it with past data, if it determines that there are no abnormalities, it will be stored in the database as is. On the other hand, if an abnormality is detected, an appropriate notification will be sent to the user and necessary contacts.

[0530] This concludes the details of the embodiment of the present invention. This system enables advanced monitoring that takes into account not only cognitive function but also the user's emotional state, improving the quality of life for the elderly and contributing to early treatment.

[0531] The processing flow will be explained below.

[0532] Step 1:

[0533] The server runs a generative model to generate everyday questions, such as "What did you have for breakfast this morning?" or "How was your day?"

[0534] Step 2:

[0535] The server transmits the generated question to the user terminal.

[0536] Step 3:

[0537] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[0538] Step 4:

[0539] The user responds naturally to questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user responds "I had bread and coffee."

[0540] Step 5:

[0541] The device records the user's response as text data and simultaneously uses an emotion engine to analyze the user's emotion when responding, for example, by recognizing emotion from voice tone.

[0542] Step 6:

[0543] The terminal transmits the collected response data and emotion data to the server.

[0544] Step 7:

[0545] The server uses the generative model to analyze the received user response data, using natural language processing techniques.

[0546] Step 8:

[0547] The server compares the response data with a database of past responses to check for consistency and unusual patterns, and also combines and analyzes the emotional data to identify any anomalies.

[0548] Step 9:

[0549] If the server detects an anomaly in the response data or emotion data as a result of the analysis, it initiates a notification process.

[0550] Step 10:

[0551] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution. For example, a message saying, "Your recent breakfast answers have been inconsistent, so please consider seeking medical advice. You may also be feeling anxious."

[0552] Step 11:

[0553] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[0554] Step 12:

[0555] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[0556] Step 13:

[0557] The terminal repeatedly asks similar questions on a daily basis and continuously collects data on the user's responses.

[0558] This system improves the accuracy of detecting abnormalities in cognitive function by taking the user's emotional state into consideration. As a specific example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning, and User A will respond with "I had tamagoyaki and rice today." If the emotion engine detects the emotion "calm," the response data and emotion data will be sent to the server. If the server's analysis reveals inconsistencies or changes in emotion, it will send an appropriate notification. In this way, continuous monitoring can be performed effortlessly in everyday life.

[0559] Example 2

[0560] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0561] As the aging society advances, early detection and treatment of dementia are becoming increasingly important. However, current methods for detecting cognitive function do not adequately reflect changes in the user's daily life or emotional state, making effective monitoring impossible. This results in low accuracy in detecting abnormalities, making it difficult to improve the user's quality of life and realize early treatment.

[0562] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0563] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for analyzing emotions contained in the answers using an emotion engine, means for detecting an abnormality based on the analysis result and emotion data, and means for notifying of detected abnormalities. This enables advanced monitoring that takes into account not only cognitive function but also emotional states.

[0564] A "generative model" is an artificial intelligence algorithm that uses natural language processing techniques to generate text for a specific task.

[0565] A "terminal" is a hardware device that can be directly operated by a user, and includes smartphones, tablets, smart speakers, etc.

[0566] A "server" is a central computer system that communicates with terminals via a network and processes and manages data.

[0567] The "means for generating questions" is a function that uses a generative model to automatically create questions to verify the user's cognitive functions.

[0568] The "means for displaying a question on a terminal" is a function for transmitting the generated question to a terminal via the Internet and presenting it to the user on that terminal in voice or text format.

[0569] The "means for collecting answers and sending them to a server" is a function for collecting answers entered by users to questions and sending them to a server via a network.

[0570] "Means for analyzing responses and determining whether there are any abnormalities in cognitive function" refers to a function that analyzes collected response data using natural language processing technology and compares it with past response history to detect abnormalities in cognitive function.

[0571] An "emotion engine" is a software component that analyzes a user's emotional state from their text data and classifies it into specific emotional categories.

[0572] "Means for analyzing emotions" is a function that uses an emotion engine to analyze the emotions contained in the user's response and uses that information in the discrimination process.

[0573] "Means for detecting abnormalities" refers to a function that detects abnormalities in cognitive function based on analysis results and emotional data.

[0574] The "notification means" is a function that notifies designated contacts such as the user, their family, or a medical institution based on detected abnormality information.

[0575] This invention is a system aimed at early detection and treatment of dementia, and is realized by combining question generation using a generative AI model, collection and analysis of user responses, and emotion recognition using an emotion engine. An embodiment of this system is described in detail below.

[0576] Question generation and display

[0577] The server runs a generative AI model to generate everyday questions to check the user's cognitive function. The generative model can be, for example, GPT-3, a large-scale natural language processing technology. For example, it generates questions such as "What did you eat this morning?" or "How was your day?" These questions are important for collecting information about the user's lifestyle habits and detecting changes in cognitive function. The generated questions are sent via the internet to the user's device. The user's device can be a smartphone, tablet, or smart speaker, and the questions are displayed in voice or text format.

[0578] Collecting user responses and analyzing sentiment

[0579] The user responds naturally to the generated questions. The user's device records the response as text data. For example, if the user responds "I ate bread and coffee," this response text is saved on the device. The device then uses an emotion engine to analyze the emotions contained in the response data. This emotion analysis can be performed using, for example, Microsoft Azure's emotion analysis API. The analysis results are output as emotion categories such as "calm," "sad," and "happy."

[0580] Data transmission and analysis

[0581] The device sends the collected response data and emotion data to the server, where it is added to a queue for analysis. The server then uses a generative model to analyze the received data. Natural language processing techniques are used for the analysis, converting the response content into a feature vector and comparing it with a database of past responses. For example, the BERT model is used to tokenize the response and generate a feature vector. Emotion data is also analyzed at the same time to determine whether there are any anomalies.

[0582] Anomaly Detection and Notification

[0583] The server detects anomalies based on the analysis results and emotional data. It compares the results with past data to determine whether there are any abnormalities in the user's cognitive function. For example, if a user has always responded "I ate coffee and bread" over the past week, always showing a calm emotion, but suddenly responds sadly, "I haven't eaten anything," this is deemed to be an abnormality in cognitive function. If an abnormality is detected, the server initiates a notification process and sends a notification to the user, their family, and registered contacts, such as medical institutions. For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution. Also, based on your emotional data, it is likely that you are feeling anxious."

[0584] Continuous data collection

[0585] The user's device continues to display similar questions on a daily basis, continuously collecting answer data and emotional data. The server receives new data, stores it in a database, and optimizes it for the next analysis. This improves the accuracy of anomaly detection and enables long-term monitoring of the user's cognitive function and emotional state.

[0586] Specific examples

[0587] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. The server analyzes the data and, if it is determined to be normal by comparing it with past data, it is stored in the database as is. If an abnormality is detected, an appropriate notification is sent to the user and necessary contacts.

[0588] This system enables advanced monitoring that takes into account not only cognitive function but also emotional state, improving the quality of life for the elderly and contributing to early treatment.

[0589] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0590] Understood. Below, we will explain the program processing flow of the dementia early detection system by dividing it into specific processing steps.

[0591] Program processing flow

[0592] Step 1:

[0593] The server uses a generative AI model to generate questions. Specifically, it uses natural language processing techniques such as GPT-3 to create everyday questions such as "What did you have for breakfast this morning?" or "How was your day?" The input is a prompt to the generative AI model, and the output is the generated question.

[0594] Step 2:

[0595] The server sends the generated question to the user's device via the Internet. The input here is the generated question, and the output is the transmission of the question to the user's device. Specifically, the server uses an HTTP request to send the question data to a smartphone or tablet.

[0596] Step 3:

[0597] The terminal displays the received question to the user in voice or text format. The input is the question sent from the server, and the output is the user's visual or auditory recognition of the question. Specifically, the text is displayed on the terminal's display, or the question is read aloud using speech synthesis.

[0598] Step 4:

[0599] The user responds naturally to the displayed question. The input is the displayed question, and the output is the user's response in text or voice. Specifically, the user can type "I ate bread and coffee" on the smartphone keyboard, or respond vocally using the voice recognition function.

[0600] Step 5:

[0601] The device records the user's responses as text data and uses an emotion engine to analyze the emotions contained in the responses. The input is the user's response data, and the output is the emotion analysis results. Specifically, the device uses Microsoft Azure's emotion analysis API to classify responses into emotion categories such as "calm," "sad," and "happy."

[0602] Step 6:

[0603] The terminal sends the collected response data and emotion data to the server. The input is response data and emotion data, and the output is data sent to the server. Specifically, an HTTP request is generated and the response data and emotion data are sent to the server in JSON format.

[0604] Step 7:

[0605] The server adds the received data to a queue for analysis and analyzes the answer using a generative model. The input is the submitted answer data and emotion data, and the output is the analysis result. Specifically, the server uses the BERT model to tokenize the answer, convert it into a feature vector, and then analyzes it.

[0606] Step 8:

[0607] The server detects anomalies based on the analysis results and emotion data. The input is the analysis results and emotion data, and the output is whether anomalies are detected. Specifically, if anomalies are found in the response patterns or emotions by comparing them with past data, a flag is set to indicate an anomaly.

[0608] Step 9:

[0609] If an anomaly is detected, the server initiates a notification process. The input is an anomaly detection flag and related information, and the output is a notification message. Specifically, a notification is sent to the user, their family, and medical institutions stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a medical institution."

[0610] Step 10:

[0611] The terminal continues to display similar questions on a daily basis and continuously collects data. The input is new questions from the server, and the output is updated answer data. Specifically, it displays newly generated questions to the user and continues to collect answers.

[0612] The above is the specific program processing flow of the dementia early detection system. Through this detailed process, it is possible to monitor the user's daily life and emotional state, detect abnormalities early, and provide appropriate notifications.

[0613] (Application example 2)

[0614] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0615] In today's world, early detection of dementia in the elderly and appropriate treatment are becoming increasingly important as the population ages. However, many elderly people do not have the opportunity to undergo regular medical checkups, making it difficult to detect cognitive decline early. Furthermore, cognitive decline is often accompanied by emotional changes, creating a need for effective monitoring methods that take this into account. Therefore, the present invention aims to provide a system that monitors changes in cognitive function in daily life and detects abnormalities early.

[0616] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0617] In this invention, the server includes means for generating questions using a generative AI model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying of detected abnormalities, means for analyzing user emotions, and means for determining abnormalities with greater accuracy based on the emotion analysis results. This enables early detection of abnormalities in cognitive function while also enabling advanced monitoring that takes emotional changes into account.

[0618] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to automatically generate questions to present to users.

[0619] A "user terminal" is a device such as a smartphone, tablet, or smart glasses that a user can directly operate to answer questions.

[0620] A "server" is a computer system that receives data sent by users, analyzes it, and stores or uses the results.

[0621] The "answer analysis means" refers to a method or device for analyzing answer data obtained from a user and determining whether there is any abnormality in cognitive function.

[0622] The "abnormality notification means" is a method or device for sending a warning to a preset notification destination when an abnormality is detected based on the analysis results.

[0623] "Emotion analysis means" refers to a method or device for analyzing a user's emotional state based on the user's responses and related data.

[0624] The "anomaly detection means" is a method or device for detecting anomalies with higher accuracy, taking into account the results of emotion analysis.

[0625] This invention relates to a system for monitoring cognitive function and recognizing emotions of elderly customers by store staff. The system generates questions using a generative AI model, displays the questions on the user's device, and collects and analyzes the user's answers. The detailed implementation of the system is described below.

[0626] Question generation and display

[0627] The server runs a generative AI model to generate cognitive function check questions for the customer. This generation process uses the "transformers" library to generate questions by inputting prompts into the GPT-2 model. For example, the prompts might look like this:

[0628] Example prompt:

[0629] "Generate questions for your customers to help detect dementia early."

[0630] The generated questions are displayed on the smartphones or smart glasses used by store staff, who then ask these questions to customers to collect everyday information.

[0631] Collecting user responses and analyzing sentiment

[0632] The user, i.e., the customer, answers questions posed by the staff. These answers are recorded on the user's device and then subjected to emotional analysis. The emotional analysis process utilizes an emotion engine API to analyze the user's emotional state from the text of their answers.

[0633] For example, if a customer answers the question "What did you eat today?" with "I had tamagoyaki and rice today," the emotion engine can detect the emotion "calm."

[0634] Data transmission and analysis

[0635] The collected response data and emotion data are sent from the user's device to a server. The server analyzes this data and determines whether there are any abnormalities in the user's cognitive function. By taking into account not only the response data but also the emotion data, the accuracy of abnormality detection is improved.

[0636] Notification and follow-up

[0637] If the server detects an abnormality as a result of the analysis, it will send a warning to staff or caregivers via an abnormality notification method. This notification will include the nature of the abnormality and its details, and provide specific advice, such as "Your answers about your recent breakfast have been inconsistent. Please consider seeking medical advice."

[0638] Continuous data collection

[0639] The user device continues to generate similar questions on a daily basis, collecting data continuously. The server accumulates the new data it receives and optimizes it for the next analysis. Accumulating emotion data can further improve the accuracy of anomaly detection.

[0640] In this way, the present invention enables store staff to effectively monitor cognitive function and recognize emotions through interactions with elderly customers, thereby enabling early detection of cognitive decline and appropriate follow-up.

[0641] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0642] Step 1: Question Generation

[0643] The server runs the generative AI model to generate questions to present to the user. In this case, it uses the prompt "Please generate questions for the customer to help with the early detection of dementia." The input is this prompt, and the output is the generated question. The generative AI model uses a transformer model such as GPT-2 to generate natural-looking questions based on the prompt.

[0644] Step 2: View the question

[0645] The terminal receives the generated question sent from the server and displays it to the user in text or audio. The input is the generated question sent from the server, and the output is the question presented to the user. For example, the question may be displayed on a smartphone or smart glasses.

[0646] Step 3: Collect responses

[0647] The user, i.e., the customer, answers questions displayed on the terminal. The input is the displayed question and the user's answer, and the output is the answer recorded in text format. The terminal records the user's answer as text data and temporarily saves it.

[0648] Step 4: Sentiment Analysis

[0649] The device sends the collected user responses to the sentiment analysis engine API, which analyzes the emotional state of the responses. The input is the user's response as text data, and the output is data indicating the emotional state. The specific operation is to send the response text to the API and receive a label such as "calm," "sad," or "happy" as the emotional state.

[0650] Step 5: Send data

[0651] The device sends the text data and the emotion analysis results to the server. The input is the user's response and the emotion analysis results, and the output is the data sent to the server. The server records the received data in a database.

[0652] Step 6: Data analysis

[0653] The server analyzes the received response data and emotional data, and compares it with past data to determine whether there are any abnormalities in cognitive function. The input is the response data and emotional data, and the output is the analysis result (normal or abnormal). Specifically, it uses natural language processing technology to convert the response content into a feature vector, which is then compared with past data to detect abnormalities.

[0654] Step 7: Notification of abnormalities

[0655] If the server detects an anomaly, it starts a notification process and notifies registered contacts that an anomaly has been detected. The input is the analysis result (anomaly) and contact information, and the output is the notification message sent. For example, a message saying "Your answers about your recent breakfast have been inconsistent, so please consider visiting a medical institution." is sent.

[0656] Step 8: Continuous data collection

[0657] The device periodically generates similar questions, collects the user's answers, and sends them to the server. The input is the new questions and the user's new answers, and the output is a continuously updated database. The server accumulates the new data and uses it for the next analysis.

[0658] The above are the processing steps for specifically implementing cognitive function monitoring and emotion recognition for elderly customers in a physical store.

[0659] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0660] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0661] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0662] [Third embodiment]

[0663] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0664] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0665] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0666] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0667] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0668] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0669] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0670] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0671] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0672] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0673] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0674] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0675] The present invention is a system aimed at early detection and treatment of dementia. This system generates questions using a generative model, presents the questions to a user via a user terminal, collects the answers, and analyzes them on a server, thereby notifying the user if any abnormalities in cognitive function are detected. The detailed implementation of this system is described below.

[0676] Question generation and display

[0677] The server uses the generative model to generate appropriate questions for the user, such as "What did you eat this morning?" or "Where did you go today?" every morning. These questions are important for gathering information about the user's lifestyle and detecting changes in cognitive function.

[0678] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[0679] Collecting and sending user responses

[0680] The user answers the displayed questions naturally. For example, the user might answer, "I ate bread and coffee." This answer is recorded as text data by the user's device. The device then transmits the collected answer data to the server.

[0681] Data Analysis and Anomaly Detection

[0682] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been responding "I ate coffee and bread" for the past week, but suddenly responds "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[0683] If an anomaly is detected based on the analysis results, the server will send a notification to the configured contacts, which can include a specific response to the anomaly and a recommended course of action.

[0684] Notification and follow-up

[0685] If an abnormality is detected, the server will notify the user, their family, and medical institutions. For example, a message such as "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[0686] Continuous data collection

[0687] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[0688] This system can promote the early detection and treatment of dementia without burdening even elderly people living alone. As a concrete example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the device will send the answer to the server, which will analyze it. If any abnormalities are detected as a result, a notification will be sent. This process makes continuous monitoring possible without any strain on the user as they go about their daily lives.

[0689] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of the elderly and contribute to early treatment.

[0690] The processing flow will be explained below.

[0691] Step 1:

[0692] The server runs the generative model to generate daily questions to check the user's cognitive functions.

[0693] Step 2:

[0694] The server transmits the generated question to the user terminal.

[0695] Step 3:

[0696] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[0697] Step 4:

[0698] The user answers questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user answers "I had bread and coffee."

[0699] Step 5:

[0700] The terminal records the user's answer as text data and transmits it to the server.

[0701] Step 6:

[0702] The server uses the generative model to analyze the received user response data, utilizing natural language processing techniques.

[0703] Step 7:

[0704] The server compares the response data with a database of past responses, checking for consistency and unusual patterns.

[0705] Step 8:

[0706] If the analysis detects an abnormality in cognitive function, the server initiates a notification process.

[0707] Step 9:

[0708] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution.

[0709] Step 10:

[0710] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[0711] Step 11:

[0712] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[0713] Step 12:

[0714] The terminal repeatedly displays questions on a daily basis and continuously transmits the user's answer data to the server.

[0715] In this way, the system continuously monitors the user's cognitive function and promptly notifies them of any abnormalities, facilitating early treatment.

[0716] Example 1

[0717] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0718] Early detection of changes in cognitive function in the elderly is important, but conventional systems require cumbersome daily data collection and analysis, making continuous monitoring difficult. Furthermore, notification methods used when abnormalities are detected are uniform, making it difficult to respond to individual circumstances. Therefore, there is a need for a system that can effectively monitor changes in cognitive function and, when abnormalities are detected, can provide prompt and appropriate responses.

[0719] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0720] In this invention, the server includes means for generating questions using a generative model, means for transmitting the generated questions to an information terminal, means for displaying the questions on the information terminal, means for collecting user answers and transmitting them to a storage device, means for analyzing the answers and determining whether there is an abnormality in cognitive function, and means for notifying the user of an abnormality based on the determination result. This makes it possible to continuously monitor the user's cognitive function in their daily life, and, if an abnormality is detected, to take prompt action according to the individual situation.

[0721] A "generative model" is a machine learning algorithm that generates appropriate questions based on prompt text.

[0722] An "information terminal" is a device that allows users to display questions and input answers, and includes smartphones, tablets, smart speakers, etc.

[0723] The "storage device" is a database or storage system that stores response data sent by users and analyzes it later.

[0724] "Analysis" refers to the process of processing the collected data using natural language processing techniques and the like to evaluate the state of the user's cognitive function.

[0725] "Abnormal" is a state that indicates a case where behavior or patterns that are different from normal are observed based on the user's response data.

[0726] "Notification" refers to the act of sending a warning or advice to the user, their family, or a medical institution when an abnormality is detected.

[0727] The present invention is a system for early detection of changes in cognitive function. This system generates appropriate questions using a generative AI model, presents the questions to a user via an information terminal, collects the answers, transmits them to a storage device, and analyzes them on a server to detect abnormalities in cognitive function. Detailed implementation methods of this system are described below.

[0728] Question generation and display

[0729] The server uses a generative model to generate appropriate questions for the user. This generative model receives a prompt as input and outputs a question based on it. An example of a prompt is "Please generate appropriate questions to ask the user to check their cognitive function." The generated questions cover various aspects of the user's daily life, such as "What did you eat this morning?" and "Where did you go today?"

[0730] The generated question is sent to an information terminal via the Internet. Information terminals are devices such as smartphones, tablets, and smart speakers, and these terminals display the question to the user by voice or text. For example, in the case of a smartphone, the message "What did you have for breakfast this morning?" is displayed on the screen, and in devices that can use voice assistants, the question is read aloud.

[0731] Collecting and sending user responses

[0732] The user answers the questions displayed or read aloud on the information terminal. For example, the user answers, "I ate bread and coffee." This answer is recorded as text data by the terminal. The recorded answer data is transmitted to a storage device via the Internet. The HTTPS protocol is used as the transmission protocol to ensure data security.

[0733] Data Analysis and Anomaly Detection

[0734] The server receives and analyzes the response data sent by the user. This analysis uses natural language processing technology. Specifically, the response content is converted into a feature vector and compared with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this could be detected as an abnormality in cognitive function.

[0735] Notification and follow-up

[0736] If an abnormality is detected based on the analysis results, the server will send a notification to the specified contacts. The notification will include the specific answers that showed the abnormality and recommended actions to take. For example, a message may be sent saying, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[0737] Continuous data collection

[0738] The information terminal continues to display similar questions on a daily basis, continuously collecting data. The server stores the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[0739] As a concrete example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the answer is sent to the server, which analyzes it. If an abnormality is detected, a notification will be sent. This process makes continuous monitoring possible without any effort in everyday life.

[0740] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of elderly people and contribute to early treatment.

[0741] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0742] Step 1: Question Generation

[0743] The server inputs a prompt to the generative model: "Generate an appropriate question to ask the user to check their cognitive function." The generative model generates a question based on this prompt. For example, the output question might be, "What did you have for breakfast this morning?" This generated question is saved in a database. The data processing performed by the server involves analyzing the prompt and generating an appropriate question.

[0744] Step 2: Submit your question

[0745] The server sends the generated question to the information terminal. At this time, a protocol for transferring data over the Internet (e.g., HTTPS) is used. The input from the server to the terminal is the generated question, and the output is the state in which the question has been sent to the terminal. The server waits for an ACK (acknowledgement) to confirm that the terminal has received the question.

[0746] Step 3: Display questions

[0747] The device receives the question from the server and displays it to the user in text or voice. For example, a message such as "What did you have for breakfast this morning?" is displayed on a smartphone screen. On devices with voice output, the question is read aloud through a voice assistant. The input to the device is the question received from the server, and the output is the question displayed to the user.

[0748] Step 4: User answers

[0749] The user inputs an answer to a question displayed on the terminal. For example, the answer may be "I ate bread and coffee." This answer can be entered as text or by voice using speech recognition. The user's input is the answer to the question, and the output is the answer entered into the terminal.

[0750] Step 5: Submit your response

[0751] The terminal records the user's answer as text data and sends it to the server. The transmission is via the Internet using the HTTPS protocol. The input to the terminal is the user's answer, and the output is the answer data sent to the server. The terminal waits for an ACK (acknowledgment) to confirm that the server has received the data.

[0752] Step 6: Data analysis

[0753] The server analyzes the received response data. Natural language processing technology is used for this analysis. The response content is converted into a feature vector and compared with a database of past responses. The server's input is the response data, and the output is the analysis result. For example, if a user who answered "I ate coffee and bread" in the past week answers "I ate nothing," the analysis result will detect an abnormality in cognitive function.

[0754] Step 7: Anomaly detection and notification

[0755] If the server detects an abnormality based on the analysis results, it will send a notification to the specified contacts. The notification will include the specific answer for the abnormality and recommended countermeasures. The server's input is the analysis results, and the output is an abnormality notification. For example, a message may be sent saying, "Your answers about breakfast are inconsistent, so please consider visiting a medical institution." An ACK (acknowledgement) is waited for to confirm that the notification was sent successfully.

[0756] Step 8: Continuous data collection

[0757] The terminal continues to display questions to the user on a daily basis. For example, it displays the question "What did you eat this morning?" every morning at 8:00. The server receives new data and stores it in the database. The terminal's input is the periodically generated question, and its output is the question displayed to the user. The server's input is the new answer data, and its output is the updated database. The database is updated and optimized for the next analysis.

[0758] (Application example 1)

[0759] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0760] It is important to detect changes in cognitive function in the elderly early and take appropriate measures promptly. However, for elderly people living alone or with busy families, daily monitoring of cognitive function is a heavy burden. In addition, manual notification methods when abnormalities are detected are time-consuming and may delay appropriate responses. It is necessary to solve these issues and improve the quality of life of the elderly.

[0761] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0762] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying the user of a detected abnormality, means for detecting an abnormality by comparing the user's past answers with the user's current answers, and means for notifying the user of a detected abnormality by email. This makes it possible to continuously monitor the user's daily life, detect abnormalities in cognitive function early, and automatically notify the user.

[0763] A "generative model" is an algorithm that generates new data based on existing data. It uses AI techniques to create new questions and answers based on specific conditions and patterns.

[0764] A "user terminal" is a device operated by a user, and includes a smartphone, tablet, or PC.

[0765] A "server" is a computer system that receives data from users over a network and processes and analyzes it.

[0766] The "means for generating questions" is the process by which the server uses the generative model to generate appropriate questions for the user.

[0767] The "means for displaying on the user terminal" is a function for displaying the generated question on the screen of the user terminal.

[0768] The "means of collecting answers" is the process of acquiring the answers users give to questions as data.

[0769] "Means for sending to the server" refers to the process of sending the response data collected from the user terminals to the server via the Internet.

[0770] "Means for analyzing responses" refers to the server's ability to verify and analyze the collected response data and identify specific patterns or anomalies.

[0771] "Means for determining whether there is any abnormality in cognitive function" is a process for evaluating whether there is any change or abnormality in the user's cognitive function based on the response data.

[0772] The "means for notifying abnormalities" is a function that sends an alert to a pre-set contact when an abnormality is detected.

[0773] "Means for comparing past answers with current answers" refers to the process of comparing a user's answer history with their most recent answers to check for consistency or anomalies.

[0774] "Means for notifying by email" is a function that automatically sends detected abnormalities to relevant parties by email.

[0775] The present invention is a system for early detection of changes in cognitive function and providing appropriate notifications when abnormalities are detected. This system generates questions using a generative model, presents the questions to the user via a user terminal, collects the answers, and analyzes them on a server, thereby providing notifications when abnormalities are detected in cognitive function.

[0776] Question generation and display

[0777] The server uses the generative model to generate appropriate questions. For example, questions such as "What did you eat this morning?" or "Where did you go today?" are generated. These questions are important for collecting information about the user's daily life. The generated questions are sent via the Internet to a user device such as a smartphone. The user device displays the questions on a screen and, in some cases, presents them to the user by voice.

[0778] Collecting and sending user responses

[0779] The user responds naturally to the displayed questions. For example, if the user responds "I ate bread and coffee," the device records the response as text data. The device then transmits the collected response data to the server.

[0780] Data Analysis and Anomaly Detection

[0781] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[0782] Abnormal notification

[0783] If an abnormality is detected based on the analysis results, the server will send a notification to pre-defined contacts. The notification will include the specific abnormality detected and recommended action to take. Notifications are automatically sent via email, allowing users, their families, and medical institutions to take action quickly.

[0784] Continuous data collection

[0785] The user's device continues to display similar questions on a daily basis, collecting data continuously. The server receives new data, stores it in a database, and optimizes it for the next analysis. This makes it possible to analyze data over a long period of time, enabling highly accurate monitoring of changes in the user's cognitive function.

[0786] Specific examples

[0787] For example, consider a case where a user answers the question "What did you eat this morning?" every morning. Suppose a user who has answered "I had bread and coffee" for a week in a row suddenly answers "I didn't eat anything." If this abnormal answer is detected, the server automatically sends an email to a pre-set contact with a message stating, "An abnormality has been detected. Your recent answers have been inconsistent." This system can constantly monitor the health of elderly people, detect abnormalities early, and promote appropriate measures.

[0788] Prompt Sentence Examples

[0789] What did you eat this morning?

[0790] What did you have for dinner last night?

[0791] Where are you planning to go today?

[0792] How are you feeling right now?

[0793] The above is an embodiment of the present invention. This system is extremely effective for efficiently monitoring the cognitive function of elderly people and promoting early treatment.

[0794] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0795] Step 1:

[0796] The server generates questions using a generative AI model.

[0797] Input: Past question data and user profile information.

[0798] Data processing: The generative AI model selects the best questions based on the conditions and generates new questions.

[0799] Output: The generated question.

[0800] Step 2:

[0801] The server transmits the generated question to the user terminal.

[0802] Input: The generated question.

[0803] Data processing: The question is converted into a data packet for transmission and sent over the Internet.

[0804] Output: The question is displayed on the user's terminal.

[0805] Step 3:

[0806] The user answers the questions displayed.

[0807] Input: The question displayed to the user.

[0808] Data processing: The user enters the answer, and the answer is recorded as text data.

[0809] Output: User answers as text data.

[0810] Step 4:

[0811] The user terminal transmits the collected response data to the server.

[0812] Input: User response text data.

[0813] Data processing: The response data is converted into a data packet for transmission and sent to the server.

[0814] Output: The response data sent to the server.

[0815] Step 5:

[0816] The server receives the response data sent by the user and begins analyzing it.

[0817] Input: The response data sent to the server.

[0818] Data processing: Using natural language processing techniques, responses are converted into feature vectors and compared with a database of past responses.

[0819] Output: Data analysis results.

[0820] Step 6:

[0821] The server compares past responses with current responses to determine whether there are any abnormalities in cognitive function.

[0822] Input: Past response database and current response data.

[0823] Data processing: Comparative analysis of the consistency of answers and abnormal patterns.

[0824] Output: Abnormality judgment result.

[0825] Step 7:

[0826] The server generates and emails notifications if any anomalies are detected.

[0827] Input: Abnormality detection result and pre-set notification destination information.

[0828] Data processing: A notification containing the details of the abnormality and countermeasures is created, converted into email format, and sent.

[0829] Output: Email notification.

[0830] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0831] This invention is a system aimed at early detection and treatment of dementia, and combines question generation using a generative model, collection and analysis of user responses, and emotion recognition using an emotion engine. The detailed implementation method of this system is described below.

[0832] Question generation and display

[0833] The server runs the generative model to generate everyday questions to check the user's cognitive function. For example, every morning it generates questions such as "What did you eat this morning?" and "How was your day?" These questions are important for collecting information about the user's lifestyle and detecting changes in cognitive function.

[0834] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[0835] Collecting and sending user responses

[0836] The user responds naturally to the displayed question. For example, suppose the user responds, "I ate bread and coffee." This response is recorded as text data by the user's device, and the emotion engine simultaneously analyzes the user's emotions. The device then transmits the collected response data and emotion data to the server.

[0837] Data Analysis and Anomaly Detection

[0838] The server receives the response data and emotion data sent by the user and analyzes the content using a generative model. Natural language processing technology is used for the analysis. The response content is converted into a feature vector and compared with a database of past responses. Meanwhile, the emotion data analyzed by the emotion engine is also added to the anomaly detection process.

[0839] For example, if a user has answered "I ate coffee and bread" over the past week and always displayed a calm emotion, but suddenly answers sadly and says "I haven't eaten anything," this would be judged to be a cognitive abnormality.

[0840] Notification and follow-up

[0841] If the analysis detects any abnormalities in cognitive function, the server initiates the notification process. It generates a notification containing the abnormality and its details and sends it to registered contacts. Recipients of the notification include the user, their family, and medical institutions.

[0842] For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a doctor. Also, based on your emotional data, it's likely that you're feeling anxious."

[0843] Continuous data collection

[0844] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. It also accumulates emotional data to improve the accuracy of anomaly detection.

[0845] Specific examples

[0846] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. After the server analyzes the data and compares it with past data, if it determines that there are no abnormalities, it will be stored in the database as is. On the other hand, if an abnormality is detected, an appropriate notification will be sent to the user and necessary contacts.

[0847] This concludes the details of the embodiment of the present invention. This system enables advanced monitoring that takes into account not only cognitive function but also the user's emotional state, improving the quality of life for the elderly and contributing to early treatment.

[0848] The processing flow will be explained below.

[0849] Step 1:

[0850] The server runs a generative model to generate everyday questions, such as "What did you have for breakfast this morning?" or "How was your day?"

[0851] Step 2:

[0852] The server transmits the generated question to the user terminal.

[0853] Step 3:

[0854] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[0855] Step 4:

[0856] The user responds naturally to questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user responds "I had bread and coffee."

[0857] Step 5:

[0858] The device records the user's response as text data and simultaneously uses an emotion engine to analyze the user's emotion when responding, for example, by recognizing emotion from voice tone.

[0859] Step 6:

[0860] The terminal transmits the collected response data and emotion data to the server.

[0861] Step 7:

[0862] The server uses the generative model to analyze the received user response data, using natural language processing techniques.

[0863] Step 8:

[0864] The server compares the response data with a database of past responses to check for consistency and unusual patterns, and also combines and analyzes the emotional data to identify any anomalies.

[0865] Step 9:

[0866] If the server detects an anomaly in the response data or emotion data as a result of the analysis, it initiates a notification process.

[0867] Step 10:

[0868] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution. For example, a message saying, "Your recent breakfast answers have been inconsistent, so please consider seeking medical advice. You may also be feeling anxious."

[0869] Step 11:

[0870] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[0871] Step 12:

[0872] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[0873] Step 13:

[0874] The terminal repeatedly asks similar questions on a daily basis and continuously collects data on the user's responses.

[0875] This system improves the accuracy of detecting abnormalities in cognitive function by taking the user's emotional state into consideration. As a specific example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning, and User A will respond with "I had tamagoyaki and rice today." If the emotion engine detects the emotion "calm," the response data and emotion data will be sent to the server. If the server's analysis reveals inconsistencies or changes in emotion, it will send an appropriate notification. In this way, continuous monitoring can be performed effortlessly in everyday life.

[0876] Example 2

[0877] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0878] As the aging society advances, early detection and treatment of dementia are becoming increasingly important. However, current methods for detecting cognitive function do not adequately reflect changes in the user's daily life or emotional state, making effective monitoring impossible. This results in low accuracy in detecting abnormalities, making it difficult to improve the user's quality of life and realize early treatment.

[0879] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0880] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for analyzing emotions contained in the answers using an emotion engine, means for detecting an abnormality based on the analysis result and emotion data, and means for notifying of detected abnormalities. This enables advanced monitoring that takes into account not only cognitive function but also emotional states.

[0881] A "generative model" is an artificial intelligence algorithm that uses natural language processing techniques to generate text for a specific task.

[0882] A "terminal" is a hardware device that can be directly operated by a user, and includes smartphones, tablets, smart speakers, etc.

[0883] A "server" is a central computer system that communicates with terminals via a network and processes and manages data.

[0884] The "means for generating questions" is a function that uses a generative model to automatically create questions to verify the user's cognitive functions.

[0885] The "means for displaying a question on a terminal" is a function for transmitting the generated question to a terminal via the Internet and presenting it to the user on that terminal in voice or text format.

[0886] The "means for collecting answers and sending them to a server" is a function for collecting answers entered by users to questions and sending them to a server via a network.

[0887] "Means for analyzing responses and determining whether there are any abnormalities in cognitive function" refers to a function that analyzes collected response data using natural language processing technology and compares it with past response history to detect abnormalities in cognitive function.

[0888] An "emotion engine" is a software component that analyzes a user's emotional state from their text data and classifies it into specific emotional categories.

[0889] "Means for analyzing emotions" is a function that uses an emotion engine to analyze the emotions contained in the user's response and uses that information in the discrimination process.

[0890] "Means for detecting abnormalities" refers to a function that detects abnormalities in cognitive function based on analysis results and emotional data.

[0891] The "notification means" is a function that notifies designated contacts such as the user, their family, or a medical institution based on detected abnormality information.

[0892] This invention is a system aimed at early detection and treatment of dementia, and is realized by combining question generation using a generative AI model, collection and analysis of user responses, and emotion recognition using an emotion engine. An embodiment of this system is described in detail below.

[0893] Question generation and display

[0894] The server runs a generative AI model to generate everyday questions to check the user's cognitive function. The generative model can be, for example, GPT-3, a large-scale natural language processing technology. For example, it generates questions such as "What did you eat this morning?" or "How was your day?" These questions are important for collecting information about the user's lifestyle habits and detecting changes in cognitive function. The generated questions are sent via the internet to the user's device. The user's device can be a smartphone, tablet, or smart speaker, and the questions are displayed in voice or text format.

[0895] Collecting user responses and analyzing sentiment

[0896] The user responds naturally to the generated questions. The user's device records the response as text data. For example, if the user responds "I ate bread and coffee," this response text is saved on the device. The device then uses an emotion engine to analyze the emotions contained in the response data. This emotion analysis can be performed using, for example, Microsoft Azure's emotion analysis API. The analysis results are output as emotion categories such as "calm," "sad," and "happy."

[0897] Data transmission and analysis

[0898] The device sends the collected response data and emotion data to the server, where it is added to a queue for analysis. The server then uses a generative model to analyze the received data. Natural language processing techniques are used for the analysis, converting the response content into a feature vector and comparing it with a database of past responses. For example, the BERT model is used to tokenize the response and generate a feature vector. Emotion data is also analyzed at the same time to determine whether there are any anomalies.

[0899] Anomaly Detection and Notification

[0900] The server detects anomalies based on the analysis results and emotional data. It compares the results with past data to determine whether there are any abnormalities in the user's cognitive function. For example, if a user has always responded "I ate coffee and bread" over the past week, always showing a calm emotion, but suddenly responds sadly, "I haven't eaten anything," this is deemed to be an abnormality in cognitive function. If an abnormality is detected, the server initiates a notification process and sends a notification to the user, their family, and registered contacts, such as medical institutions. For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution. Also, based on your emotional data, it is likely that you are feeling anxious."

[0901] Continuous data collection

[0902] The user's device continues to display similar questions on a daily basis, continuously collecting answer data and emotional data. The server receives new data, stores it in a database, and optimizes it for the next analysis. This improves the accuracy of anomaly detection and enables long-term monitoring of the user's cognitive function and emotional state.

[0903] Specific examples

[0904] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. The server analyzes the data and, if it is determined to be normal by comparing it with past data, it is stored in the database as is. If an abnormality is detected, an appropriate notification is sent to the user and necessary contacts.

[0905] This system enables advanced monitoring that takes into account not only cognitive function but also emotional state, improving the quality of life for the elderly and contributing to early treatment.

[0906] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0907] Understood. Below, we will explain the program processing flow of the dementia early detection system by dividing it into specific processing steps.

[0908] Program processing flow

[0909] Step 1:

[0910] The server uses a generative AI model to generate questions. Specifically, it uses natural language processing techniques such as GPT-3 to create everyday questions such as "What did you have for breakfast this morning?" or "How was your day?" The input is a prompt to the generative AI model, and the output is the generated question.

[0911] Step 2:

[0912] The server sends the generated question to the user's device via the Internet. The input here is the generated question, and the output is the transmission of the question to the user's device. Specifically, the server uses an HTTP request to send the question data to a smartphone or tablet.

[0913] Step 3:

[0914] The terminal displays the received question to the user in voice or text format. The input is the question sent from the server, and the output is the user's visual or auditory recognition of the question. Specifically, the text is displayed on the terminal's display, or the question is read aloud using speech synthesis.

[0915] Step 4:

[0916] The user responds naturally to the displayed question. The input is the displayed question, and the output is the user's response in text or voice. Specifically, the user can type "I ate bread and coffee" on the smartphone keyboard, or respond vocally using the voice recognition function.

[0917] Step 5:

[0918] The device records the user's responses as text data and uses an emotion engine to analyze the emotions contained in the responses. The input is the user's response data, and the output is the emotion analysis results. Specifically, the device uses Microsoft Azure's emotion analysis API to classify responses into emotion categories such as "calm," "sad," and "happy."

[0919] Step 6:

[0920] The terminal sends the collected response data and emotion data to the server. The input is response data and emotion data, and the output is data sent to the server. Specifically, an HTTP request is generated and the response data and emotion data are sent to the server in JSON format.

[0921] Step 7:

[0922] The server adds the received data to a queue for analysis and analyzes the answer using a generative model. The input is the submitted answer data and emotion data, and the output is the analysis result. Specifically, the server uses the BERT model to tokenize the answer, convert it into a feature vector, and then analyzes it.

[0923] Step 8:

[0924] The server detects anomalies based on the analysis results and emotion data. The input is the analysis results and emotion data, and the output is whether anomalies are detected. Specifically, if anomalies are found in the response patterns or emotions by comparing them with past data, a flag is set to indicate an anomaly.

[0925] Step 9:

[0926] If an anomaly is detected, the server initiates a notification process. The input is an anomaly detection flag and related information, and the output is a notification message. Specifically, a notification is sent to the user, their family, and medical institutions stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a medical institution."

[0927] Step 10:

[0928] The terminal continues to display similar questions on a daily basis and continuously collects data. The input is new questions from the server, and the output is updated answer data. Specifically, it displays newly generated questions to the user and continues to collect answers.

[0929] The above is the specific program processing flow of the dementia early detection system. Through this detailed process, it is possible to monitor the user's daily life and emotional state, detect abnormalities early, and provide appropriate notifications.

[0930] (Application example 2)

[0931] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0932] In today's world, early detection of dementia in the elderly and appropriate treatment are becoming increasingly important as the population ages. However, many elderly people do not have the opportunity to undergo regular medical checkups, making it difficult to detect cognitive decline early. Furthermore, cognitive decline is often accompanied by emotional changes, creating a need for effective monitoring methods that take this into account. Therefore, the present invention aims to provide a system that monitors changes in cognitive function in daily life and detects abnormalities early.

[0933] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0934] In this invention, the server includes means for generating questions using a generative AI model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying of detected abnormalities, means for analyzing user emotions, and means for determining abnormalities with greater accuracy based on the emotion analysis results. This enables early detection of abnormalities in cognitive function while also enabling advanced monitoring that takes emotional changes into account.

[0935] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to automatically generate questions to present to users.

[0936] A "user terminal" is a device such as a smartphone, tablet, or smart glasses that a user can directly operate to answer questions.

[0937] A "server" is a computer system that receives data sent by users, analyzes it, and stores or uses the results.

[0938] The "answer analysis means" refers to a method or device for analyzing answer data obtained from a user and determining whether there is any abnormality in cognitive function.

[0939] The "abnormality notification means" is a method or device for sending a warning to a preset notification destination when an abnormality is detected based on the analysis results.

[0940] "Emotion analysis means" refers to a method or device for analyzing a user's emotional state based on the user's responses and related data.

[0941] The "anomaly detection means" is a method or device for detecting anomalies with higher accuracy, taking into account the results of emotion analysis.

[0942] This invention relates to a system for monitoring cognitive function and recognizing emotions of elderly customers by store staff. The system generates questions using a generative AI model, displays the questions on the user's device, and collects and analyzes the user's answers. The detailed implementation of the system is described below.

[0943] Question generation and display

[0944] The server runs a generative AI model to generate cognitive function check questions for the customer. This generation process uses the "transformers" library to generate questions by inputting prompts into the GPT-2 model. For example, the prompts might look like this:

[0945] Example prompt:

[0946] "Generate questions for your customers to help detect dementia early."

[0947] The generated questions are displayed on the smartphones or smart glasses used by store staff, who then ask these questions to customers to collect everyday information.

[0948] Collecting user responses and analyzing sentiment

[0949] The user, i.e., the customer, answers questions posed by the staff. These answers are recorded on the user's device and then subjected to emotional analysis. The emotional analysis process utilizes an emotion engine API to analyze the user's emotional state from the text of their answers.

[0950] For example, if a customer answers the question "What did you eat today?" with "I had tamagoyaki and rice today," the emotion engine can detect the emotion "calm."

[0951] Data transmission and analysis

[0952] The collected response data and emotion data are sent from the user's device to a server. The server analyzes this data and determines whether there are any abnormalities in the user's cognitive function. By taking into account not only the response data but also the emotion data, the accuracy of abnormality detection is improved.

[0953] Notification and follow-up

[0954] If the server detects an abnormality as a result of the analysis, it will send a warning to staff or caregivers via an abnormality notification method. This notification will include the nature of the abnormality and its details, and provide specific advice, such as "Your answers about your recent breakfast have been inconsistent. Please consider seeking medical advice."

[0955] Continuous data collection

[0956] The user device continues to generate similar questions on a daily basis, collecting data continuously. The server accumulates the new data it receives and optimizes it for the next analysis. Accumulating emotion data can further improve the accuracy of anomaly detection.

[0957] In this way, the present invention enables store staff to effectively monitor cognitive function and recognize emotions through interactions with elderly customers, thereby enabling early detection of cognitive decline and appropriate follow-up.

[0958] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0959] Step 1: Question Generation

[0960] The server runs the generative AI model to generate questions to present to the user. In this case, it uses the prompt "Please generate questions for the customer to help with the early detection of dementia." The input is this prompt, and the output is the generated question. The generative AI model uses a transformer model such as GPT-2 to generate natural-looking questions based on the prompt.

[0961] Step 2: View the question

[0962] The terminal receives the generated question sent from the server and displays it to the user in text or audio. The input is the generated question sent from the server, and the output is the question presented to the user. For example, the question may be displayed on a smartphone or smart glasses.

[0963] Step 3: Collect responses

[0964] The user, i.e., the customer, answers questions displayed on the terminal. The input is the displayed question and the user's answer, and the output is the answer recorded in text format. The terminal records the user's answer as text data and temporarily saves it.

[0965] Step 4: Sentiment Analysis

[0966] The device sends the collected user responses to the sentiment analysis engine API, which analyzes the emotional state of the responses. The input is the user's response as text data, and the output is data indicating the emotional state. The specific operation is to send the response text to the API and receive a label such as "calm," "sad," or "happy" as the emotional state.

[0967] Step 5: Send data

[0968] The device sends the text data and the emotion analysis results to the server. The input is the user's response and the emotion analysis results, and the output is the data sent to the server. The server records the received data in a database.

[0969] Step 6: Data analysis

[0970] The server analyzes the received response data and emotional data, and compares it with past data to determine whether there are any abnormalities in cognitive function. The input is the response data and emotional data, and the output is the analysis result (normal or abnormal). Specifically, it uses natural language processing technology to convert the response content into a feature vector, which is then compared with past data to detect abnormalities.

[0971] Step 7: Notification of abnormalities

[0972] If the server detects an anomaly, it starts a notification process and notifies registered contacts that an anomaly has been detected. The input is the analysis result (anomaly) and contact information, and the output is the notification message sent. For example, a message saying "Your answers about your recent breakfast have been inconsistent, so please consider visiting a medical institution." is sent.

[0973] Step 8: Continuous data collection

[0974] The device periodically generates similar questions, collects the user's answers, and sends them to the server. The input is the new questions and the user's new answers, and the output is a continuously updated database. The server accumulates the new data and uses it for the next analysis.

[0975] The above are the processing steps for specifically implementing cognitive function monitoring and emotion recognition for elderly customers in a physical store.

[0976] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0977] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0978] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[0979] [Fourth embodiment]

[0980] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0981] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0982] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0983] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0984] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0985] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0986] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0987] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0988] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0989] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0990] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0991] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0992] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0993] The present invention is a system aimed at early detection and treatment of dementia. This system generates questions using a generative model, presents the questions to a user via a user terminal, collects the answers, and analyzes them on a server, thereby notifying the user if any abnormalities in cognitive function are detected. The detailed implementation of this system is described below.

[0994] Question generation and display

[0995] The server uses the generative model to generate appropriate questions for the user, such as "What did you eat this morning?" or "Where did you go today?" every morning. These questions are important for gathering information about the user's lifestyle and detecting changes in cognitive function.

[0996] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[0997] Collecting and sending user responses

[0998] The user answers the displayed questions naturally. For example, the user might answer, "I ate bread and coffee." This answer is recorded as text data by the user's device. The device then transmits the collected answer data to the server.

[0999] Data Analysis and Anomaly Detection

[1000] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been responding "I ate coffee and bread" for the past week, but suddenly responds "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[1001] If an anomaly is detected based on the analysis results, the server will send a notification to the configured contacts, which can include a specific response to the anomaly and a recommended course of action.

[1002] Notification and follow-up

[1003] If an abnormality is detected, the server will notify the user, their family, and medical institutions. For example, a message such as "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[1004] Continuous data collection

[1005] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[1006] This system can promote the early detection and treatment of dementia without burdening even elderly people living alone. As a concrete example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the device will send the answer to the server, which will analyze it. If any abnormalities are detected as a result, a notification will be sent. This process makes continuous monitoring possible without any strain on the user as they go about their daily lives.

[1007] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of the elderly and contribute to early treatment.

[1008] The processing flow will be explained below.

[1009] Step 1:

[1010] The server runs the generative model to generate daily questions to check the user's cognitive functions.

[1011] Step 2:

[1012] The server transmits the generated question to the user terminal.

[1013] Step 3:

[1014] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[1015] Step 4:

[1016] The user answers questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user answers "I had bread and coffee."

[1017] Step 5:

[1018] The terminal records the user's answer as text data and transmits it to the server.

[1019] Step 6:

[1020] The server uses the generative model to analyze the received user response data, utilizing natural language processing techniques.

[1021] Step 7:

[1022] The server compares the response data with a database of past responses, checking for consistency and unusual patterns.

[1023] Step 8:

[1024] If the analysis detects an abnormality in cognitive function, the server initiates a notification process.

[1025] Step 9:

[1026] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution.

[1027] Step 10:

[1028] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[1029] Step 11:

[1030] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[1031] Step 12:

[1032] The terminal repeatedly displays questions on a daily basis and continuously transmits the user's answer data to the server.

[1033] In this way, the system continuously monitors the user's cognitive function and promptly notifies them of any abnormalities, facilitating early treatment.

[1034] Example 1

[1035] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1036] Early detection of changes in cognitive function in the elderly is important, but conventional systems require cumbersome daily data collection and analysis, making continuous monitoring difficult. Furthermore, notification methods used when abnormalities are detected are uniform, making it difficult to respond to individual circumstances. Therefore, there is a need for a system that can effectively monitor changes in cognitive function and, when abnormalities are detected, can provide prompt and appropriate responses.

[1037] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1038] In this invention, the server includes means for generating questions using a generative model, means for transmitting the generated questions to an information terminal, means for displaying the questions on the information terminal, means for collecting user answers and transmitting them to a storage device, means for analyzing the answers and determining whether there is an abnormality in cognitive function, and means for notifying the user of an abnormality based on the determination result. This makes it possible to continuously monitor the user's cognitive function in their daily life, and, if an abnormality is detected, to take prompt action according to the individual situation.

[1039] A "generative model" is a machine learning algorithm that generates appropriate questions based on prompt text.

[1040] An "information terminal" is a device that allows users to display questions and input answers, and includes smartphones, tablets, smart speakers, etc.

[1041] The "storage device" is a database or storage system that stores response data sent by users and analyzes it later.

[1042] "Analysis" refers to the process of processing the collected data using natural language processing techniques and the like to evaluate the state of the user's cognitive function.

[1043] "Abnormal" is a state that indicates a case where behavior or patterns that are different from normal are observed based on the user's response data.

[1044] "Notification" refers to the act of sending a warning or advice to the user, their family, or a medical institution when an abnormality is detected.

[1045] The present invention is a system for early detection of changes in cognitive function. This system generates appropriate questions using a generative AI model, presents the questions to a user via an information terminal, collects the answers, transmits them to a storage device, and analyzes them on a server to detect abnormalities in cognitive function. Detailed implementation methods of this system are described below.

[1046] Question generation and display

[1047] The server uses a generative model to generate appropriate questions for the user. This generative model receives a prompt as input and outputs a question based on it. An example of a prompt is "Please generate appropriate questions to ask the user to check their cognitive function." The generated questions cover various aspects of the user's daily life, such as "What did you eat this morning?" and "Where did you go today?"

[1048] The generated question is sent to an information terminal via the Internet. Information terminals are devices such as smartphones, tablets, and smart speakers, and these terminals display the question to the user by voice or text. For example, in the case of a smartphone, the message "What did you have for breakfast this morning?" is displayed on the screen, and in devices that can use voice assistants, the question is read aloud.

[1049] Collecting and sending user responses

[1050] The user answers the questions displayed or read aloud on the information terminal. For example, the user answers, "I ate bread and coffee." This answer is recorded as text data by the terminal. The recorded answer data is transmitted to a storage device via the Internet. The HTTPS protocol is used as the transmission protocol to ensure data security.

[1051] Data Analysis and Anomaly Detection

[1052] The server receives and analyzes the response data sent by the user. This analysis uses natural language processing technology. Specifically, the response content is converted into a feature vector and compared with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this could be detected as an abnormality in cognitive function.

[1053] Notification and follow-up

[1054] If an abnormality is detected based on the analysis results, the server will send a notification to the specified contacts. The notification will include the specific answers that showed the abnormality and recommended actions to take. For example, a message may be sent saying, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution." This allows for early and appropriate action to be taken.

[1055] Continuous data collection

[1056] The information terminal continues to display similar questions on a daily basis, continuously collecting data. The server stores the new data it receives and optimizes it for the next analysis. This enables long-term data analysis and highly accurate monitoring of changes in the user's cognitive function.

[1057] As a concrete example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. If User A answers "I had tamagoyaki and rice today," the answer is sent to the server, which analyzes it. If an abnormality is detected, a notification will be sent. This process makes continuous monitoring possible without any effort in everyday life.

[1058] The above is a detailed description of the embodiment of the present invention. This system can improve the quality of life of elderly people and contribute to early treatment.

[1059] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1060] Step 1: Question Generation

[1061] The server inputs a prompt to the generative model: "Generate an appropriate question to ask the user to check their cognitive function." The generative model generates a question based on this prompt. For example, the output question might be, "What did you have for breakfast this morning?" This generated question is saved in a database. The data processing performed by the server involves analyzing the prompt and generating an appropriate question.

[1062] Step 2: Submit your question

[1063] The server sends the generated question to the information terminal. At this time, a protocol for transferring data over the Internet (e.g., HTTPS) is used. The input from the server to the terminal is the generated question, and the output is the state in which the question has been sent to the terminal. The server waits for an ACK (acknowledgement) to confirm that the terminal has received the question.

[1064] Step 3: Display questions

[1065] The device receives the question from the server and displays it to the user in text or voice. For example, a message such as "What did you have for breakfast this morning?" is displayed on a smartphone screen. On devices with voice output, the question is read aloud through a voice assistant. The input to the device is the question received from the server, and the output is the question displayed to the user.

[1066] Step 4: User answers

[1067] The user inputs an answer to a question displayed on the terminal. For example, the answer may be "I ate bread and coffee." This answer can be entered as text or by voice using speech recognition. The user's input is the answer to the question, and the output is the answer entered into the terminal.

[1068] Step 5: Submit your response

[1069] The terminal records the user's answer as text data and sends it to the server. The transmission is via the Internet using the HTTPS protocol. The input to the terminal is the user's answer, and the output is the answer data sent to the server. The terminal waits for an ACK (acknowledgment) to confirm that the server has received the data.

[1070] Step 6: Data analysis

[1071] The server analyzes the received response data. Natural language processing technology is used for this analysis. The response content is converted into a feature vector and compared with a database of past responses. The server's input is the response data, and the output is the analysis result. For example, if a user who answered "I ate coffee and bread" in the past week answers "I ate nothing," the analysis result will detect an abnormality in cognitive function.

[1072] Step 7: Anomaly detection and notification

[1073] If the server detects an abnormality based on the analysis results, it will send a notification to the specified contacts. The notification will include the specific answer for the abnormality and recommended countermeasures. The server's input is the analysis results, and the output is an abnormality notification. For example, a message may be sent saying, "Your answers about breakfast are inconsistent, so please consider visiting a medical institution." An ACK (acknowledgement) is waited for to confirm that the notification was sent successfully.

[1074] Step 8: Continuous data collection

[1075] The terminal continues to display questions to the user on a daily basis. For example, it displays the question "What did you eat this morning?" every morning at 8:00. The server receives new data and stores it in the database. The terminal's input is the periodically generated question, and its output is the question displayed to the user. The server's input is the new answer data, and its output is the updated database. The database is updated and optimized for the next analysis.

[1076] (Application example 1)

[1077] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1078] It is important to detect changes in cognitive function in the elderly early and take appropriate measures promptly. However, for elderly people living alone or with busy families, daily monitoring of cognitive function is a heavy burden. In addition, manual notification methods when abnormalities are detected are time-consuming and may delay appropriate responses. It is necessary to solve these issues and improve the quality of life of the elderly.

[1079] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1080] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying the user of a detected abnormality, means for detecting an abnormality by comparing the user's past answers with the user's current answers, and means for notifying the user of a detected abnormality by email. This makes it possible to continuously monitor the user's daily life, detect abnormalities in cognitive function early, and automatically notify the user.

[1081] A "generative model" is an algorithm that generates new data based on existing data. It uses AI techniques to create new questions and answers based on specific conditions and patterns.

[1082] A "user terminal" is a device operated by a user, and includes a smartphone, tablet, or PC.

[1083] A "server" is a computer system that receives data from users over a network and processes and analyzes it.

[1084] The "means for generating questions" is the process by which the server uses the generative model to generate appropriate questions for the user.

[1085] The "means for displaying on the user terminal" is a function for displaying the generated question on the screen of the user terminal.

[1086] The "means of collecting answers" is the process of acquiring the answers users give to questions as data.

[1087] "Means for sending to the server" refers to the process of sending the response data collected from the user terminals to the server via the Internet.

[1088] "Means for analyzing responses" refers to the server's ability to verify and analyze the collected response data and identify specific patterns or anomalies.

[1089] "Means for determining whether there is any abnormality in cognitive function" is a process for evaluating whether there is any change or abnormality in the user's cognitive function based on the response data.

[1090] The "means for notifying abnormalities" is a function that sends an alert to a pre-set contact when an abnormality is detected.

[1091] "Means for comparing past answers with current answers" refers to the process of comparing a user's answer history with their most recent answers to check for consistency or anomalies.

[1092] "Means for notifying by email" is a function that automatically sends detected abnormalities to relevant parties by email.

[1093] The present invention is a system for early detection of changes in cognitive function and providing appropriate notifications when abnormalities are detected. This system generates questions using a generative model, presents the questions to the user via a user terminal, collects the answers, and analyzes them on a server, thereby providing notifications when abnormalities are detected in cognitive function.

[1094] Question generation and display

[1095] The server uses the generative model to generate appropriate questions. For example, questions such as "What did you eat this morning?" or "Where did you go today?" are generated. These questions are important for collecting information about the user's daily life. The generated questions are sent via the Internet to a user device such as a smartphone. The user device displays the questions on a screen and, in some cases, presents them to the user by voice.

[1096] Collecting and sending user responses

[1097] The user responds naturally to the displayed questions. For example, if the user responds "I ate bread and coffee," the device records the response as text data. The device then transmits the collected response data to the server.

[1098] Data Analysis and Anomaly Detection

[1099] The server receives the response data sent by the user and analyzes it using a generative model. Natural language processing technology is used for the analysis, converting the response into a feature vector and comparing it with a database of past responses. For example, if a user has been answering "I ate coffee and bread" for the past week, but suddenly answers "I haven't eaten anything," this is deemed to be a cognitive abnormality.

[1100] Abnormal notification

[1101] If an abnormality is detected based on the analysis results, the server will send a notification to pre-defined contacts. The notification will include the specific abnormality detected and recommended action to take. Notifications are automatically sent via email, allowing users, their families, and medical institutions to take action quickly.

[1102] Continuous data collection

[1103] The user's device continues to display similar questions on a daily basis, collecting data continuously. The server receives new data, stores it in a database, and optimizes it for the next analysis. This makes it possible to analyze data over a long period of time, enabling highly accurate monitoring of changes in the user's cognitive function.

[1104] Specific examples

[1105] For example, consider a case where a user answers the question "What did you eat this morning?" every morning. Suppose a user who has answered "I had bread and coffee" for a week in a row suddenly answers "I didn't eat anything." If this abnormal answer is detected, the server automatically sends an email to a pre-set contact with a message stating, "An abnormality has been detected. Your recent answers have been inconsistent." This system can constantly monitor the health of elderly people, detect abnormalities early, and promote appropriate measures.

[1106] Prompt Sentence Examples

[1107] What did you eat this morning?

[1108] What did you have for dinner last night?

[1109] Where are you planning to go today?

[1110] How are you feeling right now?

[1111] The above is an embodiment of the present invention. This system is extremely effective for efficiently monitoring the cognitive function of elderly people and promoting early treatment.

[1112] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1113] Step 1:

[1114] The server generates questions using a generative AI model.

[1115] Input: Past question data and user profile information.

[1116] Data processing: The generative AI model selects the best questions based on the conditions and generates new questions.

[1117] Output: The generated question.

[1118] Step 2:

[1119] The server transmits the generated question to the user terminal.

[1120] Input: The generated question.

[1121] Data processing: The question is converted into a data packet for transmission and sent over the Internet.

[1122] Output: The question is displayed on the user's terminal.

[1123] Step 3:

[1124] The user answers the questions displayed.

[1125] Input: The question displayed to the user.

[1126] Data processing: The user enters the answer, and the answer is recorded as text data.

[1127] Output: User answers as text data.

[1128] Step 4:

[1129] The user terminal transmits the collected response data to the server.

[1130] Input: User response text data.

[1131] Data processing: The response data is converted into a data packet for transmission and sent to the server.

[1132] Output: The response data sent to the server.

[1133] Step 5:

[1134] The server receives the response data sent by the user and begins analyzing it.

[1135] Input: The response data sent to the server.

[1136] Data processing: Using natural language processing techniques, responses are converted into feature vectors and compared with a database of past responses.

[1137] Output: Data analysis results.

[1138] Step 6:

[1139] The server compares past responses with current responses to determine whether there are any abnormalities in cognitive function.

[1140] Input: Past response database and current response data.

[1141] Data processing: Comparative analysis of the consistency of answers and abnormal patterns.

[1142] Output: Abnormality judgment result.

[1143] Step 7:

[1144] The server generates and emails notifications if any anomalies are detected.

[1145] Input: Abnormality detection result and pre-set notification destination information.

[1146] Data processing: A notification containing the details of the abnormality and countermeasures is created, converted into email format, and sent.

[1147] Output: Email notification.

[1148] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1149] This invention is a system aimed at early detection and treatment of dementia, and combines question generation using a generative model, collection and analysis of user responses, and emotion recognition using an emotion engine. The detailed implementation method of this system is described below.

[1150] Question generation and display

[1151] The server runs the generative model to generate everyday questions to check the user's cognitive function. For example, every morning it generates questions such as "What did you eat this morning?" and "How was your day?" These questions are important for collecting information about the user's lifestyle and detecting changes in cognitive function.

[1152] The generated question is sent over the Internet to a user terminal, which can be a device such as a smartphone, tablet, or smart speaker, and the question is displayed to the user in voice or text format.

[1153] Collecting and sending user responses

[1154] The user responds naturally to the displayed question. For example, suppose the user responds, "I ate bread and coffee." This response is recorded as text data by the user's device, and the emotion engine simultaneously analyzes the user's emotions. The device then transmits the collected response data and emotion data to the server.

[1155] Data Analysis and Anomaly Detection

[1156] The server receives the response data and emotion data sent by the user and analyzes the content using a generative model. Natural language processing technology is used for the analysis. The response content is converted into a feature vector and compared with a database of past responses. Meanwhile, the emotion data analyzed by the emotion engine is also added to the anomaly detection process.

[1157] For example, if a user has answered "I ate coffee and bread" over the past week and always displayed a calm emotion, but suddenly answers sadly and says "I haven't eaten anything," this would be judged to be a cognitive abnormality.

[1158] Notification and follow-up

[1159] If the analysis detects any abnormalities in cognitive function, the server initiates the notification process. It generates a notification containing the abnormality and its details and sends it to registered contacts. Recipients of the notification include the user, their family, and medical institutions.

[1160] For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a doctor. Also, based on your emotional data, it's likely that you're feeling anxious."

[1161] Continuous data collection

[1162] The user's device continues to display similar questions on a daily basis, continuously collecting data. The server accumulates the new data it receives and optimizes it for the next analysis. It also accumulates emotional data to improve the accuracy of anomaly detection.

[1163] Specific examples

[1164] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. After the server analyzes the data and compares it with past data, if it determines that there are no abnormalities, it will be stored in the database as is. On the other hand, if an abnormality is detected, an appropriate notification will be sent to the user and necessary contacts.

[1165] This concludes the details of the embodiment of the present invention. This system enables advanced monitoring that takes into account not only cognitive function but also the user's emotional state, improving the quality of life for the elderly and contributing to early treatment.

[1166] The processing flow will be explained below.

[1167] Step 1:

[1168] The server runs a generative model to generate everyday questions, such as "What did you have for breakfast this morning?" or "How was your day?"

[1169] Step 2:

[1170] The server transmits the generated question to the user terminal.

[1171] Step 3:

[1172] The terminal displays the question received from the server to the user at the specified time, either in text or audio format.

[1173] Step 4:

[1174] The user responds naturally to questions displayed on the terminal. For example, in response to the question "What did you have for breakfast this morning?", the user responds "I had bread and coffee."

[1175] Step 5:

[1176] The device records the user's response as text data and simultaneously uses an emotion engine to analyze the user's emotion when responding, for example, by recognizing emotion from voice tone.

[1177] Step 6:

[1178] The terminal transmits the collected response data and emotion data to the server.

[1179] Step 7:

[1180] The server uses the generative model to analyze the received user response data, using natural language processing techniques.

[1181] Step 8:

[1182] The server compares the response data with a database of past responses to check for consistency and unusual patterns, and also combines and analyzes the emotional data to identify any anomalies.

[1183] Step 9:

[1184] If the server detects an anomaly in the response data or emotion data as a result of the analysis, it initiates a notification process.

[1185] Step 10:

[1186] The server generates a notification containing the anomaly and its details and sends it to the user, a family member, or a medical institution. For example, a message saying, "Your recent breakfast answers have been inconsistent, so please consider seeking medical advice. You may also be feeling anxious."

[1187] Step 11:

[1188] The device displays the notification received from the server to the user, using methods such as a pop-up or audio alert.

[1189] Step 12:

[1190] The server stores the analysis results and notification history in a database and prepares the data for the next analysis.

[1191] Step 13:

[1192] The terminal repeatedly asks similar questions on a daily basis and continuously collects data on the user's responses.

[1193] This system improves the accuracy of detecting abnormalities in cognitive function by taking the user's emotional state into consideration. As a specific example, if User A is using a smartphone, the device will display the question "What did you eat this morning?" at 8:00 every morning, and User A will respond with "I had tamagoyaki and rice today." If the emotion engine detects the emotion "calm," the response data and emotion data will be sent to the server. If the server's analysis reveals inconsistencies or changes in emotion, it will send an appropriate notification. In this way, continuous monitoring can be performed effortlessly in everyday life.

[1194] Example 2

[1195] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1196] As the aging society advances, early detection and treatment of dementia are becoming increasingly important. However, current methods for detecting cognitive function do not adequately reflect changes in the user's daily life or emotional state, making effective monitoring impossible. This results in low accuracy in detecting abnormalities, making it difficult to improve the user's quality of life and realize early treatment.

[1197] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1198] In this invention, the server includes means for generating questions using a generative model, means for displaying the generated questions on a terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for analyzing emotions contained in the answers using an emotion engine, means for detecting an abnormality based on the analysis result and emotion data, and means for notifying of detected abnormalities. This enables advanced monitoring that takes into account not only cognitive function but also emotional states.

[1199] A "generative model" is an artificial intelligence algorithm that uses natural language processing techniques to generate text for a specific task.

[1200] A "terminal" is a hardware device that can be directly operated by a user, and includes smartphones, tablets, smart speakers, etc.

[1201] A "server" is a central computer system that communicates with terminals via a network and processes and manages data.

[1202] The "means for generating questions" is a function that uses a generative model to automatically create questions to verify the user's cognitive functions.

[1203] The "means for displaying a question on a terminal" is a function for transmitting the generated question to a terminal via the Internet and presenting it to the user on that terminal in voice or text format.

[1204] The "means for collecting answers and sending them to a server" is a function for collecting answers entered by users to questions and sending them to a server via a network.

[1205] "Means for analyzing responses and determining whether there are any abnormalities in cognitive function" refers to a function that analyzes collected response data using natural language processing technology and compares it with past response history to detect abnormalities in cognitive function.

[1206] An "emotion engine" is a software component that analyzes a user's emotional state from their text data and classifies it into specific emotional categories.

[1207] "Means for analyzing emotions" is a function that uses an emotion engine to analyze the emotions contained in the user's response and uses that information in the discrimination process.

[1208] "Means for detecting abnormalities" refers to a function that detects abnormalities in cognitive function based on analysis results and emotional data.

[1209] The "notification means" is a function that notifies designated contacts such as the user, their family, or a medical institution based on detected abnormality information.

[1210] This invention is a system aimed at early detection and treatment of dementia, and is realized by combining question generation using a generative AI model, collection and analysis of user responses, and emotion recognition using an emotion engine. An embodiment of this system is described in detail below.

[1211] Question generation and display

[1212] The server runs a generative AI model to generate everyday questions to check the user's cognitive function. The generative model can be, for example, GPT-3, a large-scale natural language processing technology. For example, it generates questions such as "What did you eat this morning?" or "How was your day?" These questions are important for collecting information about the user's lifestyle habits and detecting changes in cognitive function. The generated questions are sent via the internet to the user's device. The user's device can be a smartphone, tablet, or smart speaker, and the questions are displayed in voice or text format.

[1213] Collecting user responses and analyzing sentiment

[1214] The user responds naturally to the generated questions. The user's device records the response as text data. For example, if the user responds "I ate bread and coffee," this response text is saved on the device. The device then uses an emotion engine to analyze the emotions contained in the response data. This emotion analysis can be performed using, for example, Microsoft Azure's emotion analysis API. The analysis results are output as emotion categories such as "calm," "sad," and "happy."

[1215] Data transmission and analysis

[1216] The device sends the collected response data and emotion data to the server, where it is added to a queue for analysis. The server then uses a generative model to analyze the received data. Natural language processing techniques are used for the analysis, converting the response content into a feature vector and comparing it with a database of past responses. For example, the BERT model is used to tokenize the response and generate a feature vector. Emotion data is also analyzed at the same time to determine whether there are any anomalies.

[1217] Anomaly Detection and Notification

[1218] The server detects anomalies based on the analysis results and emotional data. It compares the results with past data to determine whether there are any abnormalities in the user's cognitive function. For example, if a user has always responded "I ate coffee and bread" over the past week, always showing a calm emotion, but suddenly responds sadly, "I haven't eaten anything," this is deemed to be an abnormality in cognitive function. If an abnormality is detected, the server initiates a notification process and sends a notification to the user, their family, and registered contacts, such as medical institutions. For example, a message could be sent stating, "Your answers about breakfast have been inconsistent recently. Please consider visiting a medical institution. Also, based on your emotional data, it is likely that you are feeling anxious."

[1219] Continuous data collection

[1220] The user's device continues to display similar questions on a daily basis, continuously collecting answer data and emotional data. The server receives new data, stores it in a database, and optimizes it for the next analysis. This improves the accuracy of anomaly detection and enables long-term monitoring of the user's cognitive function and emotional state.

[1221] Specific examples

[1222] For example, if User A is using a smartphone, the device will display the question "What did you have for breakfast this morning?" at 8:00 every morning. User A responds, "I had tamagoyaki (rolled omelet) and rice today," and the emotion engine detects the emotion "calm." This response and emotion data are sent from the device to the server. The server analyzes the data and, if it is determined to be normal by comparing it with past data, it is stored in the database as is. If an abnormality is detected, an appropriate notification is sent to the user and necessary contacts.

[1223] This system enables advanced monitoring that takes into account not only cognitive function but also emotional state, improving the quality of life for the elderly and contributing to early treatment.

[1224] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1225] Understood. Below, we will explain the program processing flow of the dementia early detection system by dividing it into specific processing steps.

[1226] Program processing flow

[1227] Step 1:

[1228] The server uses a generative AI model to generate questions. Specifically, it uses natural language processing techniques such as GPT-3 to create everyday questions such as "What did you have for breakfast this morning?" or "How was your day?" The input is a prompt to the generative AI model, and the output is the generated question.

[1229] Step 2:

[1230] The server sends the generated question to the user's device via the Internet. The input here is the generated question, and the output is the transmission of the question to the user's device. Specifically, the server uses an HTTP request to send the question data to a smartphone or tablet.

[1231] Step 3:

[1232] The terminal displays the received question to the user in voice or text format. The input is the question sent from the server, and the output is the user's visual or auditory recognition of the question. Specifically, the text is displayed on the terminal's display, or the question is read aloud using speech synthesis.

[1233] Step 4:

[1234] The user responds naturally to the displayed question. The input is the displayed question, and the output is the user's response in text or voice. Specifically, the user can type "I ate bread and coffee" on the smartphone keyboard, or respond vocally using the voice recognition function.

[1235] Step 5:

[1236] The device records the user's responses as text data and uses an emotion engine to analyze the emotions contained in the responses. The input is the user's response data, and the output is the emotion analysis results. Specifically, the device uses Microsoft Azure's emotion analysis API to classify responses into emotion categories such as "calm," "sad," and "happy."

[1237] Step 6:

[1238] The terminal sends the collected response data and emotion data to the server. The input is response data and emotion data, and the output is data sent to the server. Specifically, an HTTP request is generated and the response data and emotion data are sent to the server in JSON format.

[1239] Step 7:

[1240] The server adds the received data to a queue for analysis and analyzes the answer using a generative model. The input is the submitted answer data and emotion data, and the output is the analysis result. Specifically, the server uses the BERT model to tokenize the answer, convert it into a feature vector, and then analyzes it.

[1241] Step 8:

[1242] The server detects anomalies based on the analysis results and emotion data. The input is the analysis results and emotion data, and the output is whether anomalies are detected. Specifically, if anomalies are found in the response patterns or emotions by comparing them with past data, a flag is set to indicate an anomaly.

[1243] Step 9:

[1244] If an anomaly is detected, the server initiates a notification process. The input is an anomaly detection flag and related information, and the output is a notification message. Specifically, a notification is sent to the user, their family, and medical institutions stating, "Your answers about breakfast have been inconsistent recently, so please consider visiting a medical institution."

[1245] Step 10:

[1246] The terminal continues to display similar questions on a daily basis and continuously collects data. The input is new questions from the server, and the output is updated answer data. Specifically, it displays newly generated questions to the user and continues to collect answers.

[1247] The above is the specific program processing flow of the dementia early detection system. Through this detailed process, it is possible to monitor the user's daily life and emotional state, detect abnormalities early, and provide appropriate notifications.

[1248] (Application example 2)

[1249] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1250] In today's world, early detection of dementia in the elderly and appropriate treatment are becoming increasingly important as the population ages. However, many elderly people do not have the opportunity to undergo regular medical checkups, making it difficult to detect cognitive decline early. Furthermore, cognitive decline is often accompanied by emotional changes, creating a need for effective monitoring methods that take this into account. Therefore, the present invention aims to provide a system that monitors changes in cognitive function in daily life and detects abnormalities early.

[1251] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1252] In this invention, the server includes means for generating questions using a generative AI model, means for displaying the generated questions on a user terminal, means for collecting user answers and sending them to the server, means for analyzing the answers and determining whether there is an abnormality in cognitive function, means for notifying of detected abnormalities, means for analyzing user emotions, and means for determining abnormalities with greater accuracy based on the emotion analysis results. This enables early detection of abnormalities in cognitive function while also enabling advanced monitoring that takes emotional changes into account.

[1253] A "generative AI model" is an artificial intelligence system that uses machine learning algorithms to automatically generate questions to present to users.

[1254] A "user terminal" is a device such as a smartphone, tablet, or smart glasses that a user can directly operate to answer questions.

[1255] A "server" is a computer system that receives data sent by users, analyzes it, and stores or uses the results.

[1256] The "answer analysis means" refers to a method or device for analyzing answer data obtained from a user and determining whether there is any abnormality in cognitive function.

[1257] The "abnormality notification means" is a method or device for sending a warning to a preset notification destination when an abnormality is detected based on the analysis results.

[1258] "Emotion analysis means" refers to a method or device for analyzing a user's emotional state based on the user's responses and related data.

[1259] The "anomaly detection means" is a method or device for detecting anomalies with higher accuracy, taking into account the results of emotion analysis.

[1260] This invention relates to a system for monitoring cognitive function and recognizing emotions of elderly customers by store staff. The system generates questions using a generative AI model, displays the questions on the user's device, and collects and analyzes the user's answers. The detailed implementation of the system is described below.

[1261] Question generation and display

[1262] The server runs a generative AI model to generate cognitive function check questions for the customer. This generation process uses the "transformers" library to generate questions by inputting prompts into the GPT-2 model. For example, the prompts might look like this:

[1263] Example prompt:

[1264] "Generate questions for your customers to help detect dementia early."

[1265] The generated questions are displayed on the smartphones or smart glasses used by store staff, who then ask these questions to customers to collect everyday information.

[1266] Collecting user responses and analyzing sentiment

[1267] The user, i.e., the customer, answers questions posed by the staff. These answers are recorded on the user's device and then subjected to emotional analysis. The emotional analysis process utilizes an emotion engine API to analyze the user's emotional state from the text of their answers.

[1268] For example, if a customer answers the question "What did you eat today?" with "I had tamagoyaki and rice today," the emotion engine can detect the emotion "calm."

[1269] Data transmission and analysis

[1270] The collected response data and emotion data are sent from the user's device to a server. The server analyzes this data and determines whether there are any abnormalities in the user's cognitive function. By taking into account not only the response data but also the emotion data, the accuracy of abnormality detection is improved.

[1271] Notification and follow-up

[1272] If the server detects an abnormality as a result of the analysis, it will send a warning to staff or caregivers via an abnormality notification method. This notification will include the nature of the abnormality and its details, and provide specific advice, such as "Your answers about your recent breakfast have been inconsistent. Please consider seeking medical advice."

[1273] Continuous data collection

[1274] The user device continues to generate similar questions on a daily basis, collecting data continuously. The server accumulates the new data it receives and optimizes it for the next analysis. Accumulating emotion data can further improve the accuracy of anomaly detection.

[1275] In this way, the present invention enables store staff to effectively monitor cognitive function and recognize emotions through interactions with elderly customers, thereby enabling early detection of cognitive decline and appropriate follow-up.

[1276] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1277] Step 1: Question Generation

[1278] The server runs the generative AI model to generate questions to present to the user. In this case, it uses the prompt "Please generate questions for the customer to help with the early detection of dementia." The input is this prompt, and the output is the generated question. The generative AI model uses a transformer model such as GPT-2 to generate natural-looking questions based on the prompt.

[1279] Step 2: View the question

[1280] The terminal receives the generated question sent from the server and displays it to the user in text or audio. The input is the generated question sent from the server, and the output is the question presented to the user. For example, the question may be displayed on a smartphone or smart glasses.

[1281] Step 3: Collect responses

[1282] The user, i.e., the customer, answers questions displayed on the terminal. The input is the displayed question and the user's answer, and the output is the answer recorded in text format. The terminal records the user's answer as text data and temporarily saves it.

[1283] Step 4: Sentiment Analysis

[1284] The device sends the collected user responses to the sentiment analysis engine API, which analyzes the emotional state of the responses. The input is the user's response as text data, and the output is data indicating the emotional state. The specific operation is to send the response text to the API and receive a label such as "calm," "sad," or "happy" as the emotional state.

[1285] Step 5: Send data

[1286] The device sends the text data and the emotion analysis results to the server. The input is the user's response and the emotion analysis results, and the output is the data sent to the server. The server records the received data in a database.

[1287] Step 6: Data analysis

[1288] The server analyzes the received response data and emotional data, and compares it with past data to determine whether there are any abnormalities in cognitive function. The input is the response data and emotional data, and the output is the analysis result (normal or abnormal). Specifically, it uses natural language processing technology to convert the response content into a feature vector, which is then compared with past data to detect abnormalities.

[1289] Step 7: Notification of abnormalities

[1290] If the server detects an anomaly, it starts a notification process and notifies registered contacts that an anomaly has been detected. The input is the analysis result (anomaly) and contact information, and the output is the notification message sent. For example, a message saying "Your answers about your recent breakfast have been inconsistent, so please consider visiting a medical institution." is sent.

[1291] Step 8: Continuous data collection

[1292] The device periodically generates similar questions, collects the user's answers, and sends them to the server. The input is the new questions and the user's new answers, and the output is a continuously updated database. The server accumulates the new data and uses it for the next analysis.

[1293] The above are the processing steps for specifically implementing cognitive function monitoring and emotion recognition for elderly customers in a physical store.

[1294] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1295] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1296] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1297] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1298] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1299] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1300] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1301] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1302] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1303] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1304] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1305] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1306] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1307] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1308] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1309] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1310] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1311] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1312] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1313] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1314] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1315] The following is further disclosed regarding the above embodiment.

[1316] (Claim 1)

[1317] a means for generating questions using the generative model;

[1318] means for displaying the generated question on a user terminal;

[1319] means for collecting and transmitting user responses to a server;

[1320] A means for analyzing the response and determining whether there is any abnormality in cognitive function;

[1321] means for notifying the detected anomaly;

[1322] A system including:

[1323] (Claim 2)

[1324] The system according to claim 1, further comprising means for enabling modification of the content of the generated question.

[1325] (Claim 3)

[1326] 2. The system according to claim 1, further comprising means for selecting a notification destination when the abnormality is detected.

[1327] (Claim 4)

[1328] 2. The system according to claim 1, wherein the analysis means comprises an algorithm for comparing a plurality of past response data to determine abnormalities.

[1329] (Claim 5)

[1330] 10. The system of claim 1, wherein the question generating means further comprises means for using different sets of questions to obtain a consistent record of the user's daily life.

[1331] (Claim 6)

[1332] 2. The system according to claim 1, wherein the user terminal comprises means for displaying and notifying the question by voice or text.

[1333] "Example 1"

[1334] (Claim 1)

[1335] a means for generating questions using the generative model;

[1336] means for transmitting the generated question to an information terminal;

[1337] means for displaying the question on the information terminal;

[1338] means for collecting and transmitting user responses to a storage device;

[1339] A means for analyzing the response and determining whether there is any abnormality in cognitive function;

[1340] a means for notifying an abnormality based on the determined result;

[1341] A system including:

[1342] (Claim 2)

[1343] The system according to claim 1, further comprising means for enabling modification of the content of the generated question.

[1344] (Claim 3)

[1345] 2. The system according to claim 1, further comprising means for selecting a notification destination when the abnormality is detected.

[1346] "Application Example 1"

[1347] (Claim 1)

[1348] a means for generating questions using the generative model;

[1349] means for displaying the generated question on a user terminal;

[1350] means for collecting and transmitting user responses to a server;

[1351] A means for analyzing the response and determining whether there is any abnormality in cognitive function;

[1352] means for notifying the detected anomaly;

[1353] means for comparing a user's past responses with a current response to detect anomalies;

[1354] a means of notifying detected anomalies by email;

[1355] A system including:

[1356] (Claim 2)

[1357] The system according to claim 1, further comprising means for enabling modification of the content of the generated question.

[1358] (Claim 3)

[1359] 2. The system according to claim 1, further comprising means for selecting a notification destination when the abnormality is detected.

[1360] "Example 2: Combining Emotion Engines"

[1361] (Claim 1)

[1362] a means for generating questions using the generative model;

[1363] means for displaying the generated question on a terminal;

[1364] means for collecting and transmitting user responses to a server;

[1365] A means for analyzing the response and determining whether there is any abnormality in cognitive function;

[1366] means for analyzing emotions contained in the responses using an emotion engine;

[1367] means for detecting an abnormality based on the analysis result and emotion data;

[1368] means for notifying the detected anomaly;

[1369] A system including:

[1370] (Claim 2)

[1371] The system according to claim 1, further comprising means for enabling modification of the content of the generated question.

[1372] (Claim 3)

[1373] 2. The system according to claim 1, further comprising means for selecting a notification destination when the abnormality is detected.

[1374] "Application example 2 when combining emotion engines"

[1375] (Claim 1)

[1376] a means for generating questions using a generative AI model;

[1377] means for displaying the generated question on a user terminal;

[1378] means for collecting and transmitting user responses to a server;

[1379] A means for analyzing the response and determining whether there is any abnormality in cognitive function;

[1380] means for notifying the detected anomaly;

[1381] means for analyzing user emotions;

[1382] A means for determining anomalies with greater accuracy based on the emotion analysis results;

[1383] A system including:

[1384] (Claim 2)

[1385] The system according to claim 1, further comprising means for enabling modification of the content of the generated question.

[1386] (Claim 3)

[1387] 2. The system according to claim 1, further comprising means for selecting a notification destination when the abnormality is detected. [Explanation of symbols]

[1388] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for generating questions using the generative model; means for displaying the generated question on a user terminal; means for collecting and transmitting user responses to a server; A means for analyzing the response and determining whether there is any abnormality in cognitive function; means for notifying the detected anomaly; A system including:

2. The system according to claim 1 , further comprising means for enabling modification of the content of the generated question.

3. The system according to claim 1 , further comprising means for selecting a notification destination when the abnormality is detected.

4. 2. The system according to claim 1, wherein the analysis means comprises an algorithm for comparing a plurality of past response data to determine abnormalities.

5. 2. The system of claim 1, wherein the question generating means further comprises means for using different sets of questions to obtain a consistent record of the user's daily life.

6. 2. The system according to claim 1, wherein the user terminal comprises means for displaying and notifying questions by voice or text.

Images